1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 self-signed certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
52 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
53 signing (which inhibits the cutthrough capability). Previously only
54 the presence of an option was tested; now an expansion evaluating as
55 empty is permissible (obviously it should depend only on data available
56 when the cutthrough connection is made).
58 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
59 the relevant preceding SMTP command did not note the pipelining mode.
61 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
62 Previously they were not counted.
64 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
65 as one having no matching records. Previously we deferred the message
66 that needed the lookup.
68 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
69 distinguished as "(=".
71 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
72 for missing MX records. Previously it only worked for missing A records.
74 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
76 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
77 after the data-go-ahead and data-ack. Patch from Jason Betts.
79 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
80 even for a "none" policy. Patch from Tony Meyer.
82 JH/22 Fix continued use of a connection for further deliveries. If a port was
83 specified by a router, it must also match for the delivery to be
86 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
87 When one of the recipients of a message was unsuitable for the connection
88 (has no matching addresses), we lost track of needing to mark it
89 deferred. As a result mail would be lost.
91 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
93 JH/25 Decoding ACL controls is now done using a binary search; the source code
94 takes up less space and should be simpler to maintain. Merge the ACL
95 condition decode tables also, with similar effect.
97 JH/26 Fix problem with one_time used on a redirect router which returned the
98 parent address unchanged. A retry would see the parent address marked as
99 delivered, so not attempt the (identical) child. As a result mail would
102 JH/27 Fix a possible security hole, wherein a process operating with the Exim
103 UID can gain a root shell. Credit to http://www.halfdog.net/ for
104 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
107 JH/28 Enable {spool,log} filesystem space and inode checks as default.
108 Main config options check_{log,spool}_{inodes,space} are now
109 100 inodes, 10MB unless set otherwise in the configuration.
111 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
112 Previously it only applied to the main-section connection policy
115 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
117 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
118 by me. Added RFC7919 DH primes as an alternative.
120 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
121 are not in the system include path.
123 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
124 GnuTLS, when a session startup failed (eg because the client disconnected)
125 Exim did stdio operations after fclose. This was exposed by a recent
126 change which nulled out the file handle after the fclose.
128 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
129 signed directly by the cert-signing cert, rather than an intermediate
130 OCSP-signing cert. This is the model used by LetsEncrypt.
132 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
137 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
138 and 3.4.4 - once the server is enabled to respond to an OCSP request
139 it does even when not requested, resulting in a stapling non-aware
140 client dropping the TLS connection.
142 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
143 support variable-length bit vectors. No functional change.
145 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
146 The I= interface field on outgoing lines is now after the H= remote
147 host field, same as incoming lines. There is a separate
148 outgoing_interface log selector which allows you to disable the
151 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
152 If not running log_selector +smtp_connection the mainlog would be held
153 open indefinitely after a "too many connections" event, including to a
154 deleted file after a log rotate. Leave the per net connection logging
155 leaving it open for efficiency as that will be quickly detected by the
156 check on the next write.
158 HS/01 Bug 1671: Fix post transport crash.
159 Processing the wait-<transport> messages could crash the delivery
160 process if the message IDs didn't exist for some reason. When
161 using 'split_spool_directory=yes' the construction of the spool
162 file name failed already, exposing the same netto behaviour.
164 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
165 mime_regex ACL conditions.
167 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
168 to DSN fail messages (bounces): remote IP, remote greeting, remote response
169 to HELO, local diagnostic string.
171 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
172 log line to debug. Even when configured with a tls authenticator many
173 client connections are expected to not authenticate in this way, so
174 an authenticate fail is not an error.
176 HS/02 Add the Exim version string to the process info. This way exiwhat
177 gives some more detail about the running daemon.
179 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
180 matter for fast-change records such as DNSBLs.
182 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
183 retry record, even if constant. There may be multiple transports with
184 different interface settings and the retry behaviour needs to be kept
187 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
189 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
191 JH/10 Bug 840: fix log_defer_output option of pipe transport
193 JH/11 Bug 830: use same host for all RCPTS of a message, even under
194 hosts_randomize. This matters a lot when combined with mua_wrapper.
196 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
197 ${quote_pgsql:<string>} operator.
199 JH/13 Bug 1708: avoid misaligned access in cached lookup.
201 JH/14 Change header file name for freeradius-client. Relevant if compiling
202 with Radius support; from the Gentoo tree and checked under Fedora.
204 JH/15 Bug 1712: Introduce $prdr_requested flag variable
206 JH/16 Bug 1714: Permit an empty string as expansion result for transport
207 option transport_filter, meaning no filtering.
209 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
211 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
212 defaults to "*" (all hosts). The variable is now available when not built
213 with TLS, default unset, mainly to enable keeping the testuite sane.
214 If a server certificate is not supplied (via tls_certificate) an error is
215 logged, and clients will find TLS connections fail on startup. Presumably
216 they will retry in-clear.
217 Packagers of Exim are strongly encouraged to create a server certificate
218 at installation time.
220 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
221 with the $config_file variable.
223 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
224 in transport context, after the attempt, and per-recipient. The latter type
225 is per host attempted. The event data is the error message, and the errno
226 information encodes the lookup type (A vs. MX) used for the (first) host,
227 and the trailing two digits of the smtp 4xx reponse.
229 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
230 to write to mainlog (or rejectlog, paniclog) in the window between file
231 creation and permissions/ownership being changed. Particularly affects
232 installations where exicyclog is run as root, rather than exim user;
233 result is that the running daemon panics and dies.
235 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
237 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
238 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
239 "pri" and "weight". Note that the previous implicit priority given by the
240 list order is no longer honoured.
242 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
245 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
246 by defining SUPPORT_SOCKS.
248 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
249 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
250 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
251 variables are renamed to proxy_{local,external}_{address,port}.
253 JH/27 Move Internationalisation support from Experimental to mainline, enabled
254 for a build by defining SUPPORT_I18N
256 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
257 of the query string, and make ${quote_redis:} do that quoting.
259 JH/29 Move Events support from Experimental to mainline, enabled by default
260 and removable for a build by defining DISABLE_EVENT.
262 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
264 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
265 cached by the daemon.
267 JH/32 Move Redis support from Experimental to mainline, enabled for a build
268 by defining LOOKUP_REDIS. The libhiredis library is required.
270 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
271 keys are given for lookup.
273 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
274 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
275 only supported when built with TLS support. The PolarSSL SHA routines
276 are still used when the TLS library is too old for convenient support.
278 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
279 openssl_options), for security. OpenSSL forces this from version 1.1.0
280 server-side so match that on older versions.
282 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
283 allocation for $value could be released as the expansion processing
284 concluded, but leaving the global pointer active for it.
286 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
287 and to use the domains and local_parts ACL conditions.
289 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
290 incorrectly not doubled on cutthrough transmission, hence seen as a
291 body-termination at the receiving system - resulting in truncated mails.
292 Commonly the sender saw a TCP-level error, and retransmitted the message
293 via the normal store-and-forward channel. This could result in duplicates
294 received - but deduplicating mailstores were liable to retain only the
295 initial truncated version.
297 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
299 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
301 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
302 we're in there, support oversigning also; bug 1309.
304 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
306 HS/04 Add support for keep_environment and add_environment options.
308 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
309 either intentional arithmetic overflow during PRNG, or testing config-
312 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
313 delivery resulted in actual delivery. Cancel cutthrough before DATA
316 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
317 rejects a recipient: pass the reject to the originator.
319 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
320 Many were false-positives and ignorable, but it's worth fixing the
323 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
324 for the new environment-manipulation done at startup. Move the routines
325 from being local to tls.c to being global via the os.c file.
327 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
328 an extract embedded as result-arg for a map, the first arg for extract
329 is unavailable so we cannot tell if this is a numbered or keyed
330 extraction. Accept either.
336 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
339 JH/02 The smtp transport option "multi_domain" is now expanded.
341 JH/03 The smtp transport now requests PRDR by default, if the server offers
344 JH/04 Certificate name checking on server certificates, when exim is a client,
345 is now done by default. The transport option tls_verify_cert_hostnames
346 can be used to disable this per-host. The build option
347 EXPERIMENTAL_CERTNAMES is withdrawn.
349 JH/05 The value of the tls_verify_certificates smtp transport and main options
350 default to the word "system" to access the system default CA bundle.
351 For GnuTLS, only version 3.0.20 or later.
353 JH/06 Verification of the server certificate for a TLS connection is now tried
354 (but not required) by default. The verification status is now logged by
355 default, for both outbound TLS and client-certificate supplying inbound
358 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
361 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
362 Status Notification (bounce) messages are now MIME format per RFC 3464.
363 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
364 under the control of the dsn_advertise_hosts option, and routers may
365 have a dsn_lasthop option.
367 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
368 default, modifiable by a malware= option. The list separator for
369 the options can now be changed in the usual way. Bug 68.
371 JH/10 The smtp_receive_timeout main option is now expanded before use.
373 JH/11 The incoming_interface log option now also enables logging of the
374 local interface on delivery outgoing connections.
376 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
377 if the interface and destination host and port all match.
379 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
382 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
383 Patch from Andrew Lewis.
385 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
386 now supports optional time-restrictions, weighting, and priority
387 modifiers per server. Patch originally by <rommer@active.by>.
389 JH/16 The spamd_address main option now supports a mixed list of local
390 and remote servers. Remote servers can be IPv6 addresses, and
391 specify a port-range.
393 JH/17 Bug 68: The spamd_address main option now supports an optional
394 timeout value per server.
396 JH/18 Bug 1581: Router and transport options headers_add/remove can
397 now have the list separator specified.
399 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
402 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
405 JH/21 Support for the A6 type of dns record is withdrawn.
407 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
408 rather than the verbs used.
410 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
411 from 255 to 1024 chars.
413 JH/24 Verification callouts now attempt to use TLS by default.
415 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
416 are generic router options now. The defaults didn't change.
418 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
419 Original patch from Alexander Shikoff, worked over by JH.
421 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
422 files if ZCAT_COMMAND is not executable.
424 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
426 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
428 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
429 Normally benign, it bites when the pair was led to by a CNAME;
430 modern usage is to not canoicalize the domain to a CNAME target
431 (and we were inconsistent anyway for A-only vs AAAA+A).
433 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
435 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
436 when evaluating $sender_host_dnssec.
438 JH/31 Check the HELO verification lookup for DNSSEC, adding new
439 $sender_helo_dnssec variable.
441 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
443 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
445 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
447 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
448 documented as working, but never had. Support all but $spam_report.
450 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
451 added for tls authenticator.
453 HS/03 Add perl_taintmode main config option
458 TL/01 When running the test suite, the README says that variables such as
459 no_msglog_check are global and can be placed anywhere in a specific
460 test's script, however it was observed that placement needed to be near
461 the beginning for it to behave that way. Changed the runtest perl
462 script to read through the entire script once to detect and set these
463 variables, reset to the beginning of the script, and then run through
464 the script parsing/test process like normal.
466 TL/02 The BSD's have an arc4random API. One of the functions to induce
467 adding randomness was arc4random_stir(), but it has been removed in
468 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
469 function when detected.
471 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
472 cause callback expansion.
474 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
475 syntax errors in an expansion can be treated as a string instead of
476 logging or causing an error, due to the internal use of bool_lax
477 instead of bool when processing it.
479 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
480 server certificates when making smtp deliveries.
482 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
484 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
486 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
488 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
489 Merged patch from Sebastian Wiedenroth.
491 JH/05 Fix results-pipe from transport process. Several recipients, combined
492 with certificate use, exposed issues where response data items split
493 over buffer boundaries were not parsed properly. This eventually
494 resulted in duplicates being sent. This issue only became common enough
495 to notice due to the introduction of conection certificate information,
496 the item size being so much larger. Found and fixed by Wolfgang Breyha.
498 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
499 size buffer was used, resulting in syntax errors when an expansion
502 JH/07 Add support for directories of certificates when compiled with a GnuTLS
503 version 3.3.6 or later.
505 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
506 is EXPERIMENTAL_EVENT, the main-configuration and transport options
507 both become "event_action", the variables become $event_name, $event_data
508 and $event_defer_errno. There is a new variable $verify_mode, usable in
509 routers, transports and related events. The tls:cert event is now also
510 raised for inbound connections, if the main configuration event_action
513 TL/06 In test suite, disable OCSP for old versions of openssl which contained
514 early OCSP support, but no stapling (appears to be less than 1.0.0).
516 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
517 server certificate names available under the smtp transport option
518 "tls_verify_cert_hostname" now do not permit multi-component wildcard
521 JH/10 Time-related extraction expansions from certificates now use the main
522 option "timezone" setting for output formatting, and are consistent
523 between OpenSSL and GnuTLS compilations. Bug 1541.
525 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
526 encoded parameter in the incoming message. Bug 1558.
528 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
529 include certificate info, eximon was claiming there were spoolfile
532 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
534 JH/14 Log delivery-related information more consistently, using the sequence
535 "H=<name> [<ip>]" wherever possible.
537 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
538 are problematic for Debian distribution, omit them from the release
541 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
543 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
546 JH/17 Fix a null-indirection in certextract expansions when a nondefault
547 output list separator was used.
552 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
553 checkers that were complaining about end of non-void function with no
556 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
557 This was a regression intruduced in 4.83 by another bugfix.
559 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
561 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
562 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
568 TF/01 Correctly close the server side of TLS when forking for delivery.
570 When a message was received over SMTP with TLS, Exim failed to clear up
571 the incoming connection properly after forking off the child process to
572 deliver the message. In some situations the subsequent outgoing
573 delivery connection happened to have the same fd number as the incoming
574 connection previously had. Exim would try to use TLS and fail, logging
575 a "Bad file descriptor" error.
577 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
578 utilities have not been installed.
580 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
581 temporary space as the ACL may create new global variables.
583 TL/01 LDAP support uses per connection or global context settings, depending
584 upon the detected version of the libraries at build time.
586 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
587 to extract and use the src ip:port in logging and expansions as if it
588 were a direct connection from the outside internet. PPv2 support was
589 updated based on HAProxy spec change in May 2014.
591 JH/02 Add ${listextract {number}{list}{success}{fail}}.
593 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
594 Properly escape header and check for NULL return.
596 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
599 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
601 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
602 characters in header names, implemented as a verify condition.
603 Contributed by Michael Fischer v. Mollard.
605 TL/05 Rename SPF condition results err_perm and err_temp to standardized
606 results permerror and temperror. Previous values are deprecated but
607 still accepted. In a future release, err_perm and err_temp will be
608 completely removed, which will be a backward incompatibility if the
609 ACL tests for either of these two old results. Patch contributed by
610 user bes-internal on the mailing list.
612 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
614 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
615 selectors, in both main and reject logs.
617 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
620 JH/07 Add malware type "sock" for talking to simple daemon.
622 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
624 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
625 routers/transports under cutthrough routing.
627 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
628 numbers. Touch up "bool" conditional to keep the same definition.
630 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
632 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
634 JH/12 Expand items in router/transport headers_add or headers_remove lists
635 individually rather than the list as a whole. Bug 1452.
637 Required for reasonable handling of multiple headers_ options when
638 they may be empty; requires that headers_remove items with embedded
639 colons must have them doubled (or the list-separator changed).
641 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
642 view the policy declared in the DMARC record. Currently, $dmarc_status
643 is a combined value of both the record presence and the result of the
646 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
648 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
649 dnslookup router and the smtp transport (applying to the forward
652 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
653 of ldap servers used for a specific lookup. Patch provided by Heiko
656 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
657 New variable $lookup_dnssec_authenticated for observability.
659 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
660 Patch submitted by Lars Timman.
662 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
664 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
665 Requires trusted mode and valid format message id, aborts otherwise.
666 Patch contributed by Heiko Schlichting.
668 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
669 certextract with support for various fields. Bug 1358.
671 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
672 is requested by default, modifiable by smtp transport option
675 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
676 operate on certificate variables to give certificate fingerprints
677 Also new ${sha256:cert_variable}.
679 JH/23 The PRDR feature is moved from being Experimental into the mainline.
681 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
682 Christian Aistleitner.
684 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
686 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
687 file. Patch from Wolfgang Breyha.
689 JH/25 Expand the coverage of the delivery $host and $host_address to
690 client authenticators run in verify callout. Bug 1476.
692 JH/26 Port service names are now accepted for tls_on_connect_ports, to
693 align with daemon_smtp_ports. Bug 72.
695 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
696 support and error reporting did not work properly.
698 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
699 and is readable. Patch from Andrew Colin Kissa.
701 TL/14 Enhance documentation of ${run expansion and how it parses the
702 commandline after expansion, particularly in the case when an
703 unquoted variable expansion results in an empty value.
705 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
707 PP/02 Fix internal collision of T_APL on systems which support RFC3123
708 by renaming away from it. Addresses GH issue 15, reported by
711 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
713 TL/15 SECURITY: prevent double expansion in math comparison functions
714 (can expand unsanitized data). Not remotely exploitable.
721 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
723 PP/02 Make -n do something, by making it not do something.
724 When combined with -bP, the name of an option is not output.
726 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
729 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
730 $sender_host_name and config options to manage this, and basic check
733 PP/05 DSCP support for outbound connections and control modifier for inbound.
735 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
736 (Only plugin which currently uses this is kerberos4, which nobody should
737 be using, but we should make it available and other future plugins might
738 conceivably use it, even though it would break NAT; stuff *should* be
739 using channel bindings instead).
741 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
742 name; added for Sendmail compatibility; requires admin caller.
743 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
744 just ignore it); requires trusted caller.
745 Also parse but ignore: -Ac -Am -X<logfile>
748 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
750 TL/02 Add +smtp_confirmation as a default logging option.
752 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
753 Patch by Magnus Holmgren from 2007-02-20.
755 TL/04 Bugzilla 1281 - Spec typo.
756 Bugzilla 1283 - Spec typo.
757 Bugzilla 1290 - Spec grammar fixes.
759 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
761 TL/06 Add Experimental DMARC support using libopendmarc libraries.
763 TL/07 Fix an out of order global option causing a segfault. Reported to dev
764 mailing list by by Dmitry Isaikin.
766 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
768 JH/02 Support "G" suffix to numbers in ${if comparisons.
770 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
772 NM/01 Bugzilla 1197 - Spec typo
773 Bugzilla 1196 - Spec examples corrections
775 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
777 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
778 gnutls_enable_pkcs11, but renamed to more accurately indicate its
781 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
782 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
784 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
785 "acl {{name}{arg}...}", and optional args on acl condition
788 JH/05 Permit multiple router/transport headers_add/remove lines.
790 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
792 JH/07 Avoid using a waiting database for a single-message-only transport.
793 Performance patch from Paul Fisher. Bugzilla 1262.
795 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
798 JH/09 Add $headers_added variable, with content from use of ACL modifier
799 add_header (but not yet added to the message). Bugzilla 199.
801 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
802 Pulled from Bugzilla 817 by Wolfgang Breyha.
804 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
806 (nb: this is the same fix as in Exim 4.80.1)
808 JH/11 Add A= logging on delivery lines, and a client_set_id option on
811 JH/12 Add optional authenticated_sender logging to A= and a log_selector
814 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
816 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
817 advertise SMTP AUTH mechanism to us, instead of a generic
818 protocol violation error. Also, make Exim more robust to bad
819 data from the Dovecot auth socket.
821 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
823 When a queue runner is handling a message, Exim first routes the
824 recipient addresses, during which it prunes them based on the retry
825 hints database. After that it attempts to deliver the message to
826 any remaining recipients. It then updates the hints database using
829 So if a recipient address works intermittently, it can get repeatedly
830 deferred at routing time. The retry hints record remains fresh so the
831 address never reaches the final cutoff time.
833 This is a fairly common occurrence when a user is bumping up against
834 their storage quota. Exim had some logic in its local delivery code
835 to deal with this. However it did not apply to per-recipient defers
836 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
838 This change adds a proper retry rule check during routing so that the
839 final cutoff time is checked against the message's age. We only do
840 this check if there is an address retry record and there is not a
841 domain retry record; this implies that previous attempts to handle
842 the address had the retry_use_local_parts option turned on. We use
843 this as an approximation for the destination being like a local
844 delivery, as in LMTP.
846 I suspect this new check makes the old local delivery cutoff check
847 redundant, but I have not verified this so I left the code in place.
849 TF/02 Correct gecos expansion when From: is a prefix of the username.
851 Test 0254 submits a message to Exim with the header
855 When I ran the test suite under the user fanf2, Exim expanded
856 the header to contain my full name, whereas it should have added
857 a Resent-Sender: header. It erroneously treats any prefix of the
858 username as equal to the username.
860 This change corrects that bug.
862 GF/01 DCC debug and logging tidyup
863 Error conditions log to paniclog rather than rejectlog.
864 Debug lines prefixed by "DCC: " to remove any ambiguity.
866 TF/03 Avoid unnecessary rebuilds of lookup-related code.
868 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
869 Bug spotted by Jeremy Harris; was flawed since initial commit.
870 Would have resulted in OCSP responses post-SNI triggering an Exim
871 NULL dereference and crash.
873 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
875 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
876 Bug detection, analysis and fix by Samuel Thibault.
877 Bugzilla 1331, Debian bug #698092.
879 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
881 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
882 Server implementation by Todd Lyons, client by JH.
883 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
884 config variable "prdr_enable" controls whether the server
885 advertises the facility. If the client requests PRDR a new
886 acl_data_smtp_prdr ACL is called once for each recipient, after
887 the body content is received and before the acl_smtp_data ACL.
888 The client is controlled by bolth of: a hosts_try_prdr option
889 on the smtp transport, and the server advertisement.
890 Default client logging of deliveries and rejections involving
891 PRDR are flagged with the string "PRDR".
893 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
894 fclose(). Diagnosis by Todd Lyons.
896 PP/17 Update configure.default to handle IPv6 localhost better.
897 Patch by Alain Williams (plus minor tweaks).
900 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
901 This is now consistent with GnuTLS, and is now documented: the
902 previous undocumented portable approach to treating the option as
903 unset was to force an expansion failure. That still works, and
904 an empty string is now equivalent.
906 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
907 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
908 not performing validation itself.
910 PP/20 Added force_command boolean option to pipe transport.
911 Patch from Nick Koston, of cPanel Inc.
913 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
916 TF/04 Added udpsend ACL modifer and hexquote expansion operator
918 PP/21 Fix eximon continuous updating with timestamped log-files.
919 Broken in a format-string cleanup in 4.80, missed when I repaired the
920 other false fix of the same issue.
921 Report and fix from Heiko Schlichting.
924 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
925 Report from Prashanth Katuri.
927 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
928 It's SecureTransport, so affects any MacOS clients which use the
929 system-integrated TLS libraries, including email clients.
931 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
932 using a MIME ACL for non-SMTP local injection.
933 Report and assistance in diagnosis by Warren Baker.
935 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
937 JH/16 Fix comparisons for 64b. Bugzilla 1385.
939 TL/09 Add expansion variable $authenticated_fail_id to keep track of
940 last id that failed so it may be referenced in subsequent ACL's.
942 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
945 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
946 ldap library initialization, allowing self-signed CA's to be
947 used. Also properly sets require_cert option later in code by
948 using NULL (global ldap config) instead of ldap handle (per
949 session). Bug diagnosis and testing by alxgomz.
951 TL/12 Enhanced documentation in the ratelimit.pl script provided in
952 the src/util/ subdirectory.
954 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
955 renamed to Transport Post Delivery Action by Jeremy Harris, as
958 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
959 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
960 redis_servers = needs to be configured which will be used by the redis
961 lookup. Patch from Warren Baker, of The Packet Hub.
963 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
965 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
966 hostname or reverse DNS when processing a host list. Used suggestions
967 from multiple comments on this bug.
969 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
971 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
972 Missed a few lines, added it to make the runtest require no keyboard
975 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
976 contains upper case chars. Make router use caseful_local_part.
978 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
979 support when GnuTLS has been built with p11-kit.
985 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
987 This, or similar/improved, will also be change PP/11 of 4.82.
993 PP/01 Handle short writes when writing local log-files.
994 In practice, only affects FreeBSD (8 onwards).
995 Bugzilla 1053, with thanks to Dmitry Isaikin.
997 NM/01 Bugzilla 949 - Documentation tweak
999 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1002 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1004 PP/02 Implemented gsasl authenticator.
1006 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1008 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1009 `pkg-config foo` for cflags/libs.
1011 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1012 with rest of GSASL and with heimdal_gssapi.
1014 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1015 `pkg-config foo` for cflags/libs for the TLS implementation.
1017 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1018 properties get this fed in as external SSF. A number of robustness
1019 and debugging improvements to the cyrus_sasl authenticator.
1021 PP/08 cyrus_sasl server now expands the server_realm option.
1023 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1024 Patch by Jeremy Harris.
1026 PP/10 Added dbmjz lookup type.
1028 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1030 PP/12 MAIL args handles TAB as well as SP, for better interop with
1031 non-compliant senders.
1032 Analysis and variant patch by Todd Lyons.
1034 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1035 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1036 Patch from Dirk Mueller <dmueller@suse.com>
1038 PP/13 tls_peerdn now print-escaped for spool files.
1039 Observed some $tls_peerdn in wild which contained \n, which resulted
1040 in spool file corruption.
1042 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1043 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1044 or write after TLS renegotiation, which otherwise led to messages
1047 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1048 as a tracking header (ie: a signed header comes before the signature).
1049 Patch from Wolfgang Breyha.
1051 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1052 comma-sep list; embedded commas doubled.
1054 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1056 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1058 Report and patch from Dmitry Banschikov.
1060 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1061 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1062 failures. We appear to now support TLS1.1+ with Exim.
1064 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1065 lets Exim select keys and certificates based upon TLS SNI from client.
1066 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1067 before an outbound SMTP session. New log_selector, +tls_sni.
1069 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1070 NULL dereference. Report and patch from Alun Jones.
1072 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1073 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1074 Not seeing resolver debug output on NetBSD, but suspect this is a
1075 resolver implementation change.
1077 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1078 Left warnings. Added "eximon gdb" invocation mode.
1080 PP/21 Defaulting "accept_8bitmime" to true, not false.
1082 PP/22 Added -bw for inetd wait mode support.
1084 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1085 locate the relevant includes and libraries. Made this the default.
1087 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1088 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1090 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1091 This may cause build issues on older platforms.
1093 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1094 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1095 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1096 Added SNI support via GnuTLS too.
1097 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1099 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1101 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1102 Applied second patch from Janne, implementing suggestion to default
1103 multiple-strings-in-record handling to match SPF spec.
1105 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1107 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1108 read-only, out of scope).
1109 Patch from Wolfgang Breyha, report from Stuart Northfield.
1111 PP/29 Fix three issues highlighted by clang analyser static analysis.
1112 Only crash-plausible issue would require the Cambridge-specific
1113 iplookup router and a misconfiguration.
1114 Report from Marcin Mirosław.
1116 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1118 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1119 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1120 As part of this, removing so much warning spew let me fix some minor
1121 real issues in debug logging.
1123 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1124 assignment on my part. Fixed.
1126 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1127 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1128 Janne Snabb (who went above and beyond: thank you).
1130 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1131 string otherwise requires a connection and a bunch more work and it's
1132 relatively easy to get wrong. Should also expose TLS library linkage
1135 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1136 64-bit ${eval} (JH/03).
1138 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1139 GNU libc to support some of the 64-bit stuff, should not lead to
1140 conflicts. Defined before os.h is pulled in, so if a given platform
1141 needs to override this, it can.
1143 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1144 protection layer was required, which is not implemented.
1145 Bugzilla 1254, patch from Wolfgang Breyha.
1147 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1148 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1149 tls_dhparam take prime identifiers. Also unbreak combination of
1150 OpenSSL+DH_params+TLSSNI.
1152 PP/39 Disable SSLv2 by default in OpenSSL support.
1158 PP/01 Solaris build fix for Oracle's LDAP libraries.
1159 Bugzilla 1109, patch from Stephen Usher.
1161 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1163 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1166 TF/02 Fix a couple more cases where we did not log the error message
1167 when unlink() failed. See also change 4.74-TF/03.
1169 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1170 lock up or crash if it happened to be inside a call to libc when it
1171 got a SIGUSR1 from exiwhat.
1173 The SIGUSR1 handler appends the current process status to the process
1174 log which is later printed by exiwhat. It used to use the general
1175 purpose logging code to do this, but several functions it calls are
1176 not safe for signals.
1178 The new output code in the SIGUSR1 handler is specific to the process
1179 log, and simple enough that it's easy to inspect for signal safety.
1180 Removing some special cases also simplifies the general logging code.
1181 Removing the spurious timestamps from the process log simplifies
1184 TF/04 Improved ratelimit ACL condition.
1186 The /noupdate option has been deprecated in favour of /readonly which
1187 has clearer semantics. The /leaky, /strict, and /readonly update modes
1188 are mutually exclusive. The update mode is no longer included in the
1189 database key; it just determines when the database is updated. (This
1190 means that when you upgrde Exim will forget old rate measurements.)
1192 Exim now checks that the per_* options are used with an update mode that
1193 makes sense for the current ACL. For example, when Exim is processing a
1194 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1195 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1196 must specify per_mail/readonly. If you omit the update mode it defaults to
1197 /leaky where that makes sense (as before) or /readonly where required.
1199 The /noupdate option is now undocumented but still supported for
1200 backwards compatibility. It is equivalent to /readonly except that in
1201 ACLs where /readonly is required you may specify /leaky/noupdate or
1202 /strict/noupdate which are treated the same as /readonly.
1204 A useful new feature is the /count= option. This is a generalization
1205 of the per_byte option, so that you can measure the throughput of other
1206 aggregate values. For example, the per_byte option is now equivalent
1207 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1209 The per_rcpt option has been generalized using the /count= mechanism
1210 (though it's more complicated than the per_byte equivalence). When it is
1211 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1212 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1213 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1214 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1215 Note that using per_rcpt with a non-readonly update mode in more than
1216 one ACL will cause the recipients to be double-counted. (The per_mail
1217 and per_byte options don't have this problem.)
1219 The handling of very low rates has changed slightly. If the computed rate
1220 is less than the event's count (usually one) then this event is the first
1221 after a long gap. In this case the rate is set to the same as this event's
1222 count, so that the first message of a spam run is counted properly.
1224 The major new feature is a mechanism for counting the rate of unique
1225 events. The new per_addr option counts the number of different
1226 recipients that someone has sent messages to in the last time period. It
1227 behaves like per_rcpt if all the recipient addresses are different, but
1228 duplicate recipient addresses do not increase the measured rate. Like
1229 the /count= option this is a general mechanism, so the per_addr option
1230 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1231 example, measure the rate that a client uses different sender addresses
1232 with the options per_mail/unique=$sender_address. There are further
1233 details in the main documentation.
1235 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1237 TF/06 Removed a few PCRE remnants.
1239 TF/07 Automatically extract Exim's version number from tags in the git
1240 repository when doing development or release builds.
1242 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1243 Bugzilla 879. Patch from Paul Fisher.
1245 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1246 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1249 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1251 PP/05 Variable $av_failed, true if the AV scanner deferred.
1252 Bugzilla 1078. Patch from John Horne.
1254 PP/06 Stop make process more reliably on build failure.
1255 Bugzilla 1087. Patch from Heiko Schlittermann.
1257 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1258 Bugzilla 1089. Patch from Heiko Schlittermann.
1260 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1261 Bugzilla 1131. Patch from Holger Weiß.
1263 PP/09 Handle IPv6 addresses with SPF.
1264 Bugzilla 860. Patch from Wolfgang Breyha.
1266 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1268 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1271 PP/11 match_* no longer expand right-hand-side by default.
1272 New compile-time build option, EXPAND_LISTMATCH_RHS.
1273 New expansion conditions, "inlist", "inlisti".
1275 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1277 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1279 PP/14 fix log_write() format string regression from TF/03.
1280 Bugzilla 1152. Patch from Dmitry Isaikin.
1286 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1288 PP/02 Harmonised TLS library version reporting; only show if debugging.
1289 Layout now matches that introduced for other libraries in 4.74 PP/03.
1291 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1293 PP/04 New "dns_use_edns0" global option.
1295 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1298 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1299 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1301 TK/01 Updated PolarSSL code to 0.14.2.
1302 Bugzilla 1097. Patch from Andreas Metzler.
1304 PP/07 Catch divide-by-zero in ${eval:...}.
1305 Fixes bugzilla 1102.
1307 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1310 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1311 format-string attack -- SECURITY: remote arbitrary code execution.
1313 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1314 time unintentionally subject to list matching rules, letting the header
1315 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1316 arbitrary string expansion). This allowed for information disclosure.
1318 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1319 INT_MIN/-1 -- value coerced to INT_MAX.
1325 NM/01 Workround for PCRE version dependency in version reporting
1328 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1329 This fixes portability to compilers other than gcc, notably
1330 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1332 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1333 makefiles for portability to HP-UX and POSIX correctness.
1335 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1336 Also via indented variable definition in the Makefile.
1337 (Debugging by Oliver Heesakkers).
1339 PP/02 Restore caching of spamd results with expanded spamd_address.
1340 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1342 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1343 Improves build reliability. Fix from: Frank Elsner
1345 NM/02 Fix wide character breakage in the rfc2047 coding
1346 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1348 NM/03 Allow underscore in dnslist lookups
1349 Fixes bug 1026. Patch from Graeme Fowler
1351 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1352 Code patches from Adam Ciarcinski of NetBSD.
1354 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1357 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1358 is logged, to avoid truncation. Patch from John Horne.
1360 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1361 Patch from Jakob Hirsch.
1363 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1364 SQL string expansion failure details.
1365 Patch from Andrey Oktyabrski.
1367 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1368 Patch from Simon Arlott.
1370 PP/09 New lookups functionality failed to compile on old gcc which rejects
1371 extern declarations in function scope.
1372 Patch from Oliver Fleischmann
1374 PP/10 Use sig_atomic_t for flags set from signal handlers.
1375 Check getgroups() return and improve debugging.
1376 Fixed developed for diagnosis in bug 927 (which turned out to be
1379 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1380 Patch from Mark Zealey.
1382 PP/12 Bugzilla 1056: Improved spamd server selection.
1383 Patch from Mark Zealey.
1385 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1386 Based on patch from Heiko Schlittermann.
1388 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1389 Patch from Uwe Doering, sign-off by Michael Haardt.
1391 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1392 variable declaration deep within a block. Bug and patch from
1395 PP/15 lookups-Makefile IRIX compatibilty coercion.
1397 PP/16 Make DISABLE_DKIM build knob functional.
1399 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1400 Patch by Simon Arlott
1402 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1403 variable argument macros. Our copy now differs from upstream.
1409 TF/01 Failure to get a lock on a hints database can have serious
1410 consequences so log it to the panic log.
1412 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1413 controlled using the smtp_confirmation log selector.
1415 TF/03 Include the error message when we fail to unlink a spool file.
1417 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1418 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1419 for maintaining out-of-tree patches for some time.
1421 PP/01 Bugzilla 139: Documentation and portability issues.
1422 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1423 Handle per-OS dynamic-module compilation flags.
1425 PP/02 Let /dev/null have normal permissions.
1426 The 4.73 fixes were a little too stringent and complained about the
1427 permissions on /dev/null. Exempt it from some checks.
1428 Reported by Andreas M. Kirchwitz.
1430 PP/03 Report version information for many libraries, including
1431 Exim version information for dynamically loaded libraries. Created
1432 version.h, now support a version extension string for distributors
1433 who patch heavily. Dynamic module ABI change.
1435 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1436 privilege escalation vulnerability whereby the Exim run-time user
1437 can cause root to append content of the attacker's choosing to
1440 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1443 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1444 If dropping privileges for untrusted macros, we disabled normal logging
1445 on the basis that it would fail; for the Exim run-time user, this is not
1446 the case, and it resulted in successful deliveries going unlogged.
1447 Fixed. Reported by Andreas Metzler.
1453 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1454 only prepend for the Resent-* case. Fixes regression introduced in
1455 Exim 4.70 by NM/22 for Bugzilla 607.
1457 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1458 increasing numbers of administrators be bitten by this.
1460 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1462 PP/03 Bugzilla 994: added openssl_options main configuration option.
1464 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1466 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1468 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1470 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1471 without a peer certificate, leading to a segfault because of an
1472 assumption that peers always have certificates. Be a little more
1473 paranoid. Problem reported by Martin Tscholak.
1475 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1476 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1477 NB: ClamAV planning to remove STREAM in "middle of 2010".
1478 CL also introduces -bmalware, various -d+acl logging additions and
1479 more caution in buffer sizes.
1481 PP/09 Implemented reverse_ip expansion operator.
1483 PP/10 Bugzilla 937: provide a "debug" ACL control.
1485 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1487 PP/12 Bugzilla 973: Implement --version.
1489 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1491 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1493 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1495 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1496 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1497 ignore trailing whitespace.
1499 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1501 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1504 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1505 Notification from Dr Andrew Aitchison.
1507 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1508 ExtendedDetectionInfo response format.
1509 Notification from John Horne.
1511 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1514 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1515 XSL and documented dependency on system catalogs, with examples of how
1518 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1521 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1522 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1523 configuration file which is writeable by the Exim user or group.
1525 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1526 of configuration files to cover files specified with the -C option if
1527 they are going to be used with root privileges, not just the default
1530 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1531 option (effectively making it always true).
1533 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1534 files to be used while preserving root privileges.
1536 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1537 that rogue child processes cannot use them.
1539 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1540 run-time user, instead of root.
1542 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1543 Exim run-time user without dropping privileges.
1545 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1546 result string, instead of calling string_vformat() twice with the same
1549 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1550 for other users. Others should always drop root privileges if they use
1551 -C on the command line, even for a whitelisted configure file.
1553 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1555 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1561 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1562 $data_path, and $header_path variables; fixed documentation bugs and
1565 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1566 exipick to access non-standard spools, including the "frozen" queue
1569 NM/01 Bugzilla 965: Support mysql stored procedures.
1570 Patch from Alain Williams
1572 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1574 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1575 Patch from Andreas Metzler
1577 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1578 Patch from Kirill Miazine
1580 NM/05 Bugzilla 671: Added umask to procmail example.
1582 JJ/03 installed exipick 20100323.0, fixing doc bug
1584 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1585 directory. Notification and patch from Dan Rosenberg.
1587 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1589 TK/02 Improve log output when DKIM signing operation fails.
1591 MH/01 Treat the transport option dkim_domain as a colon separated
1592 list, not as a single string, and sign the message with each element,
1593 omitting multiple occurences of the same signer.
1595 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1596 Bugzilla 985, 986. Patch by Simon Arlott
1598 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1599 Patch by Simon Arlott
1601 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1602 MBX locking. Notification from Dan Rosenberg.
1608 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1610 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1612 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1614 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1616 NM/04 Bugzilla 915: Flag broken perl installation during build.
1622 TK/01 Added patch by Johannes Berg that expands the main option
1623 "spamd_address" if it starts with a dollar sign.
1625 TK/02 Write list of recipients to X-Envelope-Sender header when building
1626 the mbox-format spool file for content scanning (suggested by Jakob
1629 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1630 (http://www.dcc-servers.net/) support via dccifd. Activated by
1631 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1633 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1634 by Mark Daniel Reidel <mr@df.eu>.
1636 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1637 When building exim an external PCRE library is now needed -
1638 PCRE is a system library on the majority of modern systems.
1639 See entry on PCRE_LIBS in EDITME file.
1641 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1642 conversation. Added nologin parameter to request.
1643 Patch contributed by Kirill Miazine.
1645 TF/01 Do not log submission mode rewrites if they do not change the address.
1647 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1649 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1650 log files in place. Contributed by Roberto Lima.
1652 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1654 TF/03 Bugzilla 615: When checking the local_parts router precondition
1655 after a local_part_suffix or local_part_prefix option, Exim now
1656 does not use the address's named list lookup cache, since this
1657 contains cached lookups for the whole local part.
1659 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1660 Robert Millan. Documentation is in experimental-spec.txt.
1662 TF/04 Bugzilla 668: Fix parallel build (make -j).
1664 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1666 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1667 Patch provided by Jan Srzednicki.
1669 TF/05 Leading white space used to be stripped from $spam_report which
1670 wrecked the formatting. Now it is preserved.
1672 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1673 that they are available at delivery time.
1675 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1677 TF/08 TLS error reporting now respects the incoming_interface and
1678 incoming_port log selectors.
1680 TF/09 Produce a more useful error message if an SMTP transport's hosts
1681 setting expands to an empty string.
1683 NM/06 Bugzilla 744: EXPN did not work under TLS.
1684 Patch provided by Phil Pennock.
1686 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1687 Patch provided by Richard Godbee.
1689 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1690 acl_smtp_notquit, added index entry.
1692 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1693 Patch provided by Eugene Bujak.
1695 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1696 accept(). Patch provided by Maxim Dounin.
1698 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1699 Patch provided by Phil Pennock.
1701 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1703 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1704 Patch provided by Brad "anomie" Jorsch.
1706 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1707 Patch provided by Dean Brooks.
1709 TK/05 Add native DKIM support (does not depend on external libraries).
1711 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1712 Patch provided by Graeme Fowler.
1714 NM/16 Bugzilla 851: Documentation example syntax fix.
1716 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1718 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1721 NM/19 Bugzilla 745: TLS version reporting.
1722 Patch provided by Phil Pennock.
1724 NM/20 Bugzilla 167: bool: condition support.
1725 Patch provided by Phil Pennock.
1727 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1728 clients. Patch provided by Phil Pennock.
1730 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1731 Patch provided by Brad "anomie" Jorsch.
1733 NM/23 Bugzilla 687: Fix misparses in eximstats.
1734 Patch provided by Heiko Schlittermann.
1736 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1737 Patch provided by Heiko Schlittermann.
1739 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1740 plus update to original patch.
1742 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1744 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1745 Patch provided by David Brownlee.
1747 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1749 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1751 NM/30 Bugzilla 888: TLS documentation bugfixes.
1753 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1755 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1756 Unlike the original bugzilla I have changed all shell scripts in src tree.
1758 NM/33 Bugzilla 898: Transport filter timeout fix.
1759 Patch by Todd Rinaldo.
1761 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1762 Patch by Serge Demonchaux.
1764 NM/35 Bugzilla 39: Base64 decode bug fixes.
1765 Patch by Jakob Hirsch.
1767 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1769 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1771 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1773 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1779 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1780 ALT-N's libdkim that I have put here:
1781 http://duncanthrax.net/exim-experimental/
1783 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1784 were called 'true' and it seems that C99 defines that as a reserved
1785 keyword to be used with 'bool' variable types. That means you could
1786 not include C99-style headers which use bools without triggering
1787 build errors in sieve.c.
1789 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1790 as mailq or other aliases. Changed the --help handling significantly
1791 to do whats expected. exim_usage() emits usage/help information.
1793 SC/01 Added the -bylocaldomain option to eximstats.
1795 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1797 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1799 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1805 PH/01 Another patch from the Sieve maintainer.
1807 PH/02 When an IPv6 address is converted to a string for single-key lookup
1808 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1809 dots are used instead of colons so that keys in lsearch files need not
1810 contain colons. This was done some time before quoting was made available
1811 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1812 (notated using the quote facility) so as to distinguish them from IPv4
1813 keys. This meant that lookups for IP addresses in host lists did not work
1814 for iplsearch lookups.
1816 This has been fixed by arranging for IPv6 addresses to be expressed with
1817 colons if the lookup type is iplsearch. This is not incompatible, because
1818 previously such lookups could never work.
1820 The situation is now rather anomolous, since one *can* have colons in
1821 ordinary lsearch keys. However, making the change in all cases is
1822 incompatible and would probably break a number of configurations.
1824 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1827 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1828 conversion specification without a maximum field width, thereby enabling
1829 a rogue spamd server to cause a buffer overflow. While nobody in their
1830 right mind would setup Exim to query an untrusted spamd server, an
1831 attacker that gains access to a server running spamd could potentially
1832 exploit this vulnerability to run arbitrary code as the Exim user.
1834 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1835 $primary_hostname instead of what libspf2 thinks the hosts name is.
1837 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1838 a directory entry by the name of the lookup key. Previously, if a
1839 symlink pointed to a non-existing file or a file in a directory that
1840 Exim lacked permissions to read, a lookup for a key matching that
1841 symlink would fail. Now it is enough that a matching directory entry
1842 exists, symlink or not. (Bugzilla 503.)
1844 PH/03 The body_linecount and body_zerocount variables are now exported in the
1847 PH/04 Added the $dnslist_matched variable.
1849 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1850 This means they are set thereafter only if the connection becomes
1853 PH/06 Added the client_condition to authenticators so that some can be skipped
1854 by clients under certain conditions.
1856 PH/07 The error message for a badly-placed control=no_multiline_responses left
1857 "_responses" off the end of the name.
1859 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1861 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1862 (without spaces) instead of just copying the configuration text.
1864 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1866 PH/11 Added $max_received_linelength.
1868 PH/12 Added +ignore_defer and +include_defer to host lists.
1870 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1871 way in which PCRE > 7.0 is built.
1873 PH/14 Implemented queue_only_load_latch.
1875 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1876 MAIL command. The effect was to mangle the value on 64-bit systems.
1878 PH/16 Another patch from the Sieve maintainer.
1880 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1882 PH/18 If a system quota error occurred while trying to create the file for
1883 a maildir delivery, the message "Mailbox is full" was not appended to the
1884 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1885 only to a quota excession during the actual writing of the file.
1887 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1888 characters?) which causes problems in log lines. The DN values are now
1889 passed through string_printing() before being added to log lines.
1891 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1892 and InterBase are left for another time.)
1894 PH/21 Added message_body_newlines option.
1896 PH/22 Guard against possible overflow in moan_check_errorcopy().
1898 PH/23 POSIX allows open() to be a macro; guard against that.
1900 PH/24 If the recipient of an error message contained an @ in the local part
1901 (suitably quoted, of course), incorrect values were put in $domain and
1902 $local_part during the evaluation of errors_copy.
1908 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1909 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1912 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1913 issue a MAIL command.
1915 PH/02 In an ACL statement such as
1917 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1919 if a client was not listed at all, or was listed with a value other than
1920 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1921 the condition was not true (as it should be), so access was not denied.
1922 The bug was that the ! inversion was incorrectly passed on to the second
1923 item. This has been fixed.
1925 PH/03 Added additional dnslists conditions == and =& which are different from
1926 = and & when the dns lookup returns more than one IP address.
1928 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1929 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1931 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1932 FSYNC, which compiles an option called disable_fsync that allows for
1933 bypassing fsync(). The documentation is heavily laced with warnings.
1935 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1937 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1938 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1939 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1940 including adding "make clean"; (3) Added -fPIC when compiling the test
1941 dynamically loaded module, to get rid of a warning.
1943 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1944 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1945 = 0s. The bug is otherwise harmless.
1947 PH/07 There was a bug in the dovecot authenticator such that the value of
1948 $auth1 could be overwritten, and so not correctly preserved, after a
1949 successful authentication. This usually meant that the value preserved by
1950 the server_setid option was incorrect.
1952 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1954 PH/09 Installed PCRE release 7.0.
1956 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1957 run for batched SMTP input. It is now run at the start of every message
1958 in the batch. While fixing this I discovered that the process information
1959 (output by running exiwhat) was not always getting set for -bs and -bS
1960 input. This is fixed, and it now also says "batched" for BSMTP.
1962 PH/11 Added control=no_pipelining.
1964 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1965 patch, slightly modified), and move the expansion of helo_data till after
1966 the connection is made in the smtp transport (so it can use these
1969 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1971 PH/14 Added log_selector = +pid.
1973 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1975 PH/16 Add ${if forany and ${if forall.
1977 PH/17 Added dsn_from option to vary the From: line in DSNs.
1979 PH/18 Flush SMTP output before performing a callout, unless control =
1980 no_callout_flush is set.
1982 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1983 was true (the default) a successful delivery failed to delete the retry
1984 item, thus causing premature timeout of the address. The bug is now
1987 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1989 PH/21 Long custom messages for fakedefer and fakereject are now split up
1990 into multiline reponses in the same way that messages for "deny" and
1991 other ACL rejections are.
1993 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1994 with slight modification.
1996 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1997 draft, changing the syntax and factoring some duplicate code".
1999 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2000 for deliveries of the second and subsequent messages over the same SMTP
2003 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2004 ${reduce, with only minor "tidies".
2006 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2008 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2009 expansion side effects.
2011 PH/27 When a message times out after an over-quota error from an Exim-imposed
2012 quota, the bounce message says "mailbox is full". This message was not
2013 being given when it was a system quota that was exceeded. It now should
2016 MH/03 Made $recipients available in local_scan(). local_scan() already has
2017 better access to the recipient list through recipients_list[], but
2018 $recipients can be useful in postmaster-provided expansion strings.
2020 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2021 in the case of a MAIL command with additional options following the
2022 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2023 were accidentally chopped off.
2025 PH/29 SMTP synchronization checks are implemented when a command is read -
2026 there is a check that no more input is waiting when there shouldn't be
2027 any. However, for some commands, a delay in an ACL can mean that it is
2028 some time before the response is written. In this time, more input might
2029 arrive, invalidly. So now there are extra checks after an ACL has run for
2030 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2031 pipelining has not been advertised.
2033 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2035 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2036 returned in the SMTP response when smtp_return_error_details was false.
2037 This has been fixed.
2039 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2040 instead of the C I/O that was originally supplied, because problems were
2041 reported on Solaris.
2043 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2044 Exim which did not show up earlier: it was assuming that a call to
2045 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2046 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2047 was a macro that became an assignment, so it seemed to work. This has
2048 changed to a proper function call with a void return, hence the compile
2049 error. Exim's code has been fixed.
2051 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2054 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2056 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2058 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2059 because it turns out to be too big or there is a timeout) there is no
2060 "Completed" line in the log. When some messages of this type were
2061 selected by exigrep, they were listed as "not completed". Others were
2062 picked up by some special patterns. I have improved the selection
2063 criteria to be more general.
2065 PH/38 The host_find_failed option in the manualroute router can now be set
2066 to "ignore", to completely ignore a host whose IP address cannot be
2067 found. If all hosts are ignored, the behaviour is controlled by the new
2068 host_all_ignored option.
2070 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2071 homing or because of multiple MX records with /mx) generated more than
2072 one IP address, and the following item turned out to be the local host,
2073 all the secondary addresses of the first item were incorrectly removed
2074 from the list, along with the local host and any following hosts (which
2075 is what is supposed to happen).
2077 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2078 whoever called Exim into the -H file. In the case of the daemon it was
2079 behaving confusingly. When first started, it used values for whoever
2080 started the daemon, but after a SIGHUP it used the Exim user (because it
2081 calls itself on a restart). I have changed the code so that it now always
2084 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2085 message are rejected with the same error (e.g. no authentication or bad
2086 sender address), and a DATA command is nevertheless sent (as can happen
2087 with PIPELINING or a stupid MUA), the error message that was given to the
2088 RCPT commands is included in the rejection of the DATA command. This is
2089 intended to be helpful for MUAs that show only the final error to their
2092 PH/42 Another patch from the Sieve maintainer.
2094 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2095 Eximstats - Fixed some broken HTML links and added missing column headers
2097 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2098 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2100 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2102 PH/43 Yet another patch from the Sieve maintainer.
2104 PH/44 I found a way to check for a TCP/IP connection going away before sending
2105 the response to the final '.' that terminates a message, but only in the
2106 case where the client has not sent further data following the '.'
2107 (unfortunately, this is allowed). However, in many cases there won't be
2108 any further data because there won't be any more messages to send. A call
2109 to select() can be used: if it shows that the input is "ready", there is
2110 either input waiting, or the socket has been closed. An attempt to read
2111 the next input character can distinguish the two cases. Previously, Exim
2112 would have sent an OK response which the client would never have see.
2113 This could lead to message repetition. This fix should cure that, at
2114 least in a lot of common cases.
2116 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2117 advertised in response to EHLO.
2123 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2124 fixed by 4.65/MH/01 (is this a record?) are fixed:
2126 (i) An empty string was always treated as zero by the numeric comparison
2127 operators. This behaviour has been restored.
2129 (ii) It is documented that the numeric comparison operators always treat
2130 their arguments as decimal numbers. This was broken in that numbers
2131 starting with 0 were being interpreted as octal.
2133 While fixing these problems I realized that there was another issue that
2134 hadn't been noticed. Values of message_size_limit (both the global option
2135 and the transport option) were treated as octal if they started with 0.
2136 The documentation was vague. These values are now always treated as
2137 decimal, and I will make that clear in the documentation.
2143 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2144 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2147 MH/01 Don't check that the operands of numeric comparison operators are
2148 integers when their expansion is in "skipping" mode (fixes bug
2149 introduced by 4.64-PH/07).
2151 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2152 child addresses, Exim now panics and dies. Previously, because the count
2153 is held in a short int, deliveries were likely to be lost. As such a
2154 large number of recipients for a single message is ridiculous
2155 (performance will be very, very poor), I have chosen to impose a limit
2156 rather than extend the field.
2162 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2163 leftover -K file (the existence of which was triggered by #402).
2164 While we were at it, introduced process PID as part of the -K
2165 filename. This should rule out race conditions when creating
2168 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2169 processing considerably. Previous code took too long for large mails,
2170 triggering a timeout which in turn triggers #401.
2172 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2173 in the DK code in transports.c. sendfile() is not really portable,
2174 hence the _LINUX specificness.
2176 TF/01 In the add_headers option to the mail command in an Exim filter,
2177 there was a bug that Exim would claim a syntax error in any
2178 header after the first one which had an odd number of characters
2181 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2182 callout verification, Exim cached a "reject" for the entire domain. This
2183 is correct for most verifications, but it is not correct for a recipient
2184 verification with use_sender or use_postmaster set, because in that case
2185 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2186 case of MAIL FROM:<> rejection from other early rejections (e.g.
2187 rejection of HELO). When verifying a recipient using a non-null MAIL
2188 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2189 Whatever the result of the callout, the value of the domain cache is
2190 left unchanged (for any other kind of callout, getting as far as trying
2191 RCPT means that the domain itself is ok).
2193 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2196 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2197 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2198 session, the daemon ignores ECONNECT errors and logs others; it now
2199 ignores EPIPE as well.
2201 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2202 (quoted-printable decoding).
2204 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2205 later the small subsequent patch to fix an introduced bug.
2207 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2209 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2211 PH/08 An error is now given if message_size_limit is specified negative.
2213 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2214 to be given (somewhat) arbitrary names.
2216 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2219 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2220 miscellaneous code fixes
2222 PH/10 Added the log_reject_target ACL modifier to specify where to log
2225 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2226 hostname. This is wrong, because it relates to the incoming message (and
2227 probably the interface on which it is arriving) and not to the outgoing
2228 callout (which could be using a different interface). This has been
2229 changed to use the value of the helo_data option from the smtp transport
2230 instead - this is what is used when a message is actually being sent. If
2231 there is no remote transport (possible with a router that sets up host
2232 addresses), $smtp_active_hostname is used.
2234 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2235 tweaks were necessary in order to get it to work (see also 21 below):
2236 (a) The code assumed that strncpy() returns a negative number on buffer
2237 overflow, which isn't the case. Replaced with Exim's string_format()
2239 (b) There were several signed/unsigned issues. I just did the minimum
2240 hacking in of casts. There is scope for a larger refactoring.
2241 (c) The code used strcasecmp() which is not a standard C function.
2242 Replaced with Exim's strcmpic() function.
2243 (d) The code set only $1; it now sets $auth1 as well.
2244 (e) A simple test gave the error "authentication client didn't specify
2245 service in request". It would seem that Dovecot has changed its
2246 interface. Fortunately there's a specification; I followed it and
2247 changed what the client sends and it appears to be working now.
2249 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2252 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2253 address A is aliased to B and C, where B exists and C does not. Without
2254 -v the output is "A verified" because verification stops after a
2255 successful redirection if more than one address is generated. However,
2256 with -v the child addresses are also verified. Exim was outputting "A
2257 failed to verify" and then showing the successful verification for C,
2258 with its parentage. It now outputs "B failed to verify", showing B's
2259 parentage before showing the successful verification of C.
2261 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2262 look up a TXT record in a specific list after matching in a combined
2265 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2266 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2267 they consult the DNS. I had assumed they would set it the way they
2268 wanted; and indeed my experiments on Linux seem to show that in some
2269 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2270 To be on the safe side, however, I have now made the interface to
2271 host_find_byname() similar to host_find_bydns(), with an argument
2272 containing the DNS resolver options. The host_find_byname() function now
2273 sets these options at its start, just as host_find_bydns() does. The smtp
2274 transport options dns_qualify_single and dns_search_parents are passed to
2275 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2276 of host_find_byname() use the default settings of RES_DEFNAMES
2277 (qualify_single) but not RES_DNSRCH (search_parents).
2279 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2280 spool_read_header() do less string testing, by means of a preliminary
2281 switch on the second character of optional "-foo" lines. (This is
2282 overdue, caused by the large number of possibilities that now exist.
2283 Originally there were few.) While I was there, I also converted the
2284 str(n)cmp tests so they don't re-test the leading "-" and the first
2285 character, in the hope this might squeeze out yet more improvement.
2287 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2288 flag allowing group syntax was set by the header_syntax check but not
2289 turned off, possible causing trouble later; (2) The flag was not being
2290 set at all for the header_verify test, causing "group"-style headers to
2291 be rejected. I have now set it in this case, and also caused header_
2292 verify to ignore an empty address taken from a group. While doing this, I
2293 came across some other cases where the code for allowing group syntax
2294 while scanning a header line wasn't quite right (mostly, not resetting
2295 the flag correctly in the right place). These bugs could have caused
2296 trouble for malformed header lines. I hope it is now all correct.
2298 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2299 with the "reply" argument non-NULL. The code, however (which originally
2300 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2301 but it didn't always do it. This confused somebody who was copying the
2302 code for some other use. I have removed all the tests.
2304 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2305 feature that was used to support insecure browsers during the U.S. crypto
2306 embargo. It requires special client support, and Exim is probably the
2307 only MTA that supported it -- and would never use it because real RSA is
2308 always available. This code has been removed, because it had the bad
2309 effect of slowing Exim down by computing (never used) parameters for the
2310 RSA_EXPORT functionality.
2312 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2313 authenticator to fail if there's a tab character in the incoming data
2314 (there should never be unless someone is messing about, as it's supposed
2315 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2316 if the connection is using TLS or if the remote IP is the same as the
2317 local IP, and the "valid-client-cert option" if a client certificate has
2320 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2321 authenticators. This can be used for authorization after authentication
2322 succeeds. (In the case of plaintext, it servers for both authentication
2325 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2326 if any retry times were supplied.
2328 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2329 connection, where there is no client IP address to check. In this
2330 situation, the verify now always succeeds.
2332 PH/25 Applied John Jetmore's -Mset patch.
2334 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2336 PH/27 In a string expansion for a processed (not raw) header when multiple
2337 headers of the same name were present, leading whitespace was being
2338 removed from all of them, but trailing whitespace was being removed only
2339 from the last one. Now trailing whitespace is removed from each header
2340 before concatenation. Completely empty headers in a concatenation (as
2341 before) are ignored.
2343 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2344 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2346 PH/29 [Removed. This was a change that I later backed out, and forgot to
2347 correct the ChangeLog entry (that I had efficiently created) before
2348 committing the later change.]
2350 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2351 address errors (4xx response to RCPT) over the same connection as other
2352 messages routed to the same hosts. Such deliveries are always "forced",
2353 so retry times are not inspected. This resulted in far too many retries
2354 for the affected addresses. The effect occurred only when there were more
2355 hosts than the hosts_max_try setting in the smtp transport when it had
2356 the 4xx errors. Those hosts that it had tried were not added to the list
2357 of hosts for which the message was waiting, so if all were tried, there
2358 was no problem. Two fixes have been applied:
2360 (i) If there are any address or message errors in an SMTP delivery, none
2361 of the hosts (tried or untried) are now added to the list of hosts
2362 for which the message is waiting, so the message should not be a
2363 candidate for sending over the same connection that was used for a
2364 successful delivery of some other message. This seems entirely
2365 reasonable: after all the message is NOT "waiting for some host".
2366 This is so "obvious" that I'm not sure why it wasn't done
2367 previously. Hope I haven't missed anything, but it can't do any
2368 harm, as the worst effect is to miss an optimization.
2370 (ii) If, despite (i), such a delivery is accidentally attempted, the
2371 routing retry time is respected, so at least it doesn't keep
2372 hammering the server.
2374 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2375 in ${readsocket because some servers need this prod.
2377 PH/32 Added some extra debug output when updating a wait-xxx database.
2379 PH/33 The hint "could be header name not terminated by colon", which has been
2380 given for certain expansion errors for a long time, was not being given
2381 for the ${if def:h_colon_omitted{... case.
2383 PH/34 The spec says: "With one important exception, whenever a domain list is
2384 being scanned, $domain contains the subject domain." There was at least
2385 one case where this was not true.
2387 PH/35 The error "getsockname() failed: connection reset by peer" was being
2388 written to the panic log as well as the main log, but it isn't really
2389 panic-worthy as it just means the connection died rather early on. I have
2390 removed the panic log writing for the ECONNRESET error when getsockname()
2393 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2394 runs only) independently of the message's sender address. This meant
2395 that, if the 4xx error was in fact related to the sender, a different
2396 message to the same recipient with a different sender could confuse
2397 things. In particualar, this can happen when sending to a greylisting
2398 server, but other circumstances could also provoke similar problems.
2399 I have changed the default so that the retry time for these errors is now
2400 based a combination of the sender and recipient addresses. This change
2401 can be overridden by setting address_retry_include_sender=false in the
2404 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2405 remote server are returned as part of bounce messages. This was not
2406 happening for LMTP over a pipe (the lmtp transport), but now it is the
2407 same for both kinds of LMTP.
2409 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2410 in header lines that were in fact CNAMEs. This is no longer the case.
2412 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2413 and queue runs started by the daemon processed all messages. This has
2414 been fixed so that -R and -S can now usefully be given with -q<time>.
2416 PH/40 Import PCRE release 6.7 (fixes some bugs).
2418 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2420 PH/42 Give an error if -q is specified more than once.
2422 PH/43 Renamed the variables $interface_address and $interface_port as
2423 $received_ip_address and $received_port, to make it clear that these
2424 values apply to message reception, and not to the outgoing interface when
2425 a message is delivered. (The old names remain recognized, of course.)
2427 PH/44 There was no timeout on the connect() call when using a Unix domain
2428 socket in the ${readsocket expansion. There now is.
2430 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2431 be meaningful with "accept".
2433 SC/01 Eximstats V1.43
2434 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2436 SC/02 Eximstats V1.44
2437 Use a glob alias rather than an array ref in the generated
2438 parser. This improves both readability and performance.
2440 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2441 Collect SpamAssassin and rejection statistics.
2442 Don't display local sender or destination tables unless
2443 there is data to show.
2444 Added average volumes into the top table text output.
2446 SC/04 Eximstats V1.46
2447 Collect data on the number of addresses (recipients)
2448 as well as the number of messages.
2450 SC/05 Eximstats V1.47
2451 Added 'Message too big' to the list of mail rejection
2452 reasons (thanks to Marco Gaiarin).
2454 SC/06 Eximstats V1.48
2455 Mainlog lines which have GMT offsets and are too short to
2456 have a flag are now skipped.
2458 SC/07 Eximstats V1.49 (Alain Williams)
2459 Added the -emptyok flag.
2461 SC/08 Eximstats V1.50
2462 Fixes for obtaining the IP address from reject messages.
2464 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2465 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2466 whitesspace changes from 4.64-PH/27
2468 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2471 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2472 are found, allow negative numbers in numeric criteria)
2474 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2476 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2477 to match changes made in 4.64-PH/43
2479 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2481 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2482 "domain" is actually an IP address, and force a failure. This locks out
2483 those revolvers/nameservers that support "A-for-A" lookups, in
2484 contravention of the specifications.
2486 PH/48 When a host name was looked up from an IP address, and the subsequent
2487 forward lookup of the name timed out, the host name was left in
2488 $sender_host_name, contrary to the specification.
2490 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2491 restricted to single-key lookups, Exim was not diagnosing an error if
2492 * or *@ was used with a query-style lookup.
2494 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2496 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2497 long ago, but noone interested enough thought of it. Let's just say that
2498 the "1.1" means that there are some new functions that weren't there at
2499 some point in the past.
2501 PH/51 Error processing for expansion failure of helo_data from an smtp
2502 transport during callout processing was broken.
2504 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2505 tested/used via the -bh/-bhc/-bs options.
2507 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2508 bug, fixed in subsequent PCRE releases).
2510 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2511 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2513 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2519 SC/01 Use a glob alias rather than an array ref in eximstats generated
2520 parser. This improves both readability and performance.
2522 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2523 Don't display local sender or destination tables in eximstats unless
2524 there is data to show.
2525 Added average volumes into the eximstats top table text output.
2527 SC/03 Collect data on the number of addresses (recipients) as well
2528 as the number of messages in eximstats.
2530 TF/01 Correct an error in the documentation for the redirect router. Exim
2531 does (usually) call initgroups() when daemonizing.
2533 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2534 with consistent privilege compared to when running as a daemon.
2536 TF/03 Note in the spec that $authenticated_id is not set for local
2537 submissions from trusted users.
2539 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2540 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2542 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2543 by adding some example configuration directives to the default
2544 configuration file. A little bit of work is required to uncomment the
2545 directives and define how usernames and passwords are checked, but
2546 there is now a framework to start from.
2548 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2549 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2550 without this. I don't know how relevant this is to other LDAP libraries.
2552 PH/02 Add the verb name to the "unknown ACL verb" error.
2554 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2556 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2558 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2559 directory not expanded when it should be if an expanded home directory
2560 was set for the address (which is overridden by the transport).
2562 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2565 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2566 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2567 because it is too late at that time, and has no effect.
2569 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2570 security issue with \' (bugzilla #107). I could not use the
2571 PQescapeStringConn() function, because it needs a PGconn value as one of
2574 PH/08 When testing addresses using -bt, indicate those final addresses that
2575 are duplicates that would not cause an additional delivery. At least one
2576 person was confused, thinking that -bt output corresponded to deliveries.
2577 (Suppressing duplicates isn't a good idea as you lose the information
2578 about possibly different redirections that led to the duplicates.)
2580 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2581 systems where poll() doesn't work, in particular OS X.
2583 PH/10 Added more information to debugging output for retry time not reached.
2585 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2586 operations in malware.c.
2588 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2591 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2592 syslog_duplication set false, log lines that would normally be written
2593 both the the main log and to the reject log were not written to syslog at
2596 PH/14 In the default configuration, change the use of "message" in ACL warn
2597 statements to "add_header".
2599 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2600 not followed by a command (e.g. "seen endif").
2602 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2603 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2606 PH/17 Added extra conditions to the default value of delay_warning_condition
2610 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2611 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2612 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2615 The Auto-Submitted: and various List- headers are standardised, whereas I
2616 don't think Precedence: ever was.
2618 PH/18 Refactored debugging code in route_finduser() to show more information,
2619 in particular, the error code if getpwnam() issues one.
2621 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2622 This is apparently needed in addition to the PH/07 change above to avoid
2623 any possible encoding problems.
2625 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2626 but not after initializing Perl.
2628 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2629 output them only if debugging. By default they are written stderr,
2630 apparently, which is not desirable.
2632 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2635 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2638 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2640 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2641 authenticated or an ident call has been made. Suppress the default
2642 values for $authenticated_id and $authenticated_sender (but permit -oMai
2643 and -oMas) when testing with -bh.
2645 PH/24 Re-jigged the order of the tests in the default configuration so that the
2646 tests for valid domains and recipients precede the DNS black list and CSA
2647 tests, on the grounds that those ones are more expensive.
2649 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2650 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2651 This bug exists in every version of Exim that I still have, right back to
2654 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2655 However, an attempt to turn on case-sensitivity in a regex key by
2656 including (?-i) didn't work because the subject string was already
2657 lowercased, and the effects were non-intuitive. It turns out that a
2658 one-line patch can be used to allow (?-i) to work as expected.
2664 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2665 other effects) broke the use of negated acl sub-conditions.
2667 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2670 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2671 "Deny" causes Exim to reject the incoming connection with a 554 error.
2672 Unfortunately, if there is a major crisis, such as a disk failure,
2673 tcp-wrappers gives "deny", whereas what one would like would be some
2674 kind of temporary error. A kludge has been added to help with this.
2675 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2676 554 error is used if errno is still zero or contains ENOENT (which occurs
2677 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2680 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2682 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2683 errors. Otherwise a message that provokes a temporary error (when other
2684 messages do not) can cause a whole host to time out.
2686 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2687 addresses were routed directly to files or pipes from a redirect router.
2688 File deliveries just didn't batch; pipe deliveries might have suffered
2691 PH/06 A failure to get a lock for a hints database would erroneously always say
2692 "Failed to get write lock", even when it was really a read lock.
2694 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2695 of 0600. This has been changed to use the value of the lockfile_mode
2696 option (which defaults to 0600).
2698 PH/08 Applied small patch from the Sieve maintainer.
2700 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2701 folder from quota calculations, a direct delivery into this folder messed
2702 up the contents of the maildirsize file. This was because the regex was
2703 used only to exclude .Trash (or whatever) when the size of the mailbox
2704 was calculated. There was no check that a delivery was happening into an
2705 excluded directory. This bug has been fixed by ignoring all quota
2706 processing for deliveries into excluded directories.
2708 PH/10 Added the maildirfolder_create_regex option to appendfile.
2714 PH/01 The code for finding all the local interface addresses on a FreeBSD
2715 system running IPv6 was broken. This may well have applied to all BSD
2716 systems, as well as to others that have similar system calls. The broken
2717 code found IPv4 interfaces correctly, but gave incorrect values for the
2718 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2719 that it would not match correctly against @[] and not recognize the IPv6
2722 PH/02 The ipliteral router was not recognizing addresses of the form user@
2723 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2725 PH/03 Added disable_ipv6.
2727 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2728 decoded form, because it is most often used to construct To: headers
2729 lines in autoreplies, and the decoded form may well be syntactically
2730 invalid. However, $reply_address has leading white space removed, and all
2731 newlines turned into spaces so that the autoreply transport does not
2734 PH/05 If group was specified without a user on a router, and no group or user
2735 was specified on a transport, the group from the router was ignored.
2737 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2738 for visible compile-time settings that can be used to change these
2739 numbers, for those that want even more. Backwards compatibility with old
2740 spool files has been maintained. However, going back to a previous Exim
2741 release will lost any variables that are in spool files.
2743 PH/07 Two small changes when running in the test harness: increase delay when
2744 passing a TCP/IP connection to a new process, in case the original
2745 process has to generate a bounce, and remove special handling of
2746 127.0.0.2 (sic), which is no longer necessary.
2748 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2749 be the same on different OS.
2751 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2754 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2755 whether --show-vars was specified or not
2757 JJ/02 exipick: Added support for new ACL variable spool format introduced
2760 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2761 syntactically invalid From: or Reply-to: line, and a filter used this to
2762 generate an autoreply, and therefore failed to obtain an address for the
2763 autoreply, Exim could try to deliver to a non-existent relative file
2764 name, causing unrelated and misleading errors. What now happens is that
2765 it logs this as a hard delivery error, but does not attempt to create a
2768 PH/11 The exinext utility has a -C option for testing purposes, but although
2769 the given file was scanned by exinext itself; it wasn't being passed on
2770 when Exim was called.
2772 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2773 an end-of-file indication when reading a command response.
2775 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2776 compiled. In many other places in Exim, IPv6 addresses are always
2777 recognized, so I have changed this. It also means that IPv4 domain
2778 literals of the form [IPV4:n.n.n.n] are now always recognized.
2780 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2781 used if the router is not running as root, for example, when verifying at
2782 ACL time, or when using -bh. The debugging output from this situation was
2783 non-existent - all you got was a failure to exec. I have made two
2786 (a) Failures to set uid/gid, the current directory, or a process leader
2787 in a subprocess such as that created by queryprogram now generate
2788 suitable debugging ouput when -d is set.
2790 (b) The queryprogram router detects when it is not running as root,
2791 outputs suitable debugging information if -d is set, and then runs
2792 the subprocess without attempting to change uid/gid.
2794 PH/15 Minor change to Makefile for building test_host (undocumented testing
2797 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2798 additional section of a DNS packet that returns MX or SRV records.
2799 Instead, it always explicitly searches for A/AAAA records. This avoids
2800 major problems that occur when a DNS server includes only records of one
2801 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2802 fixed another bug: if SRV records were looked up and the corresponding
2803 address records were *not* found in the additional section, the port
2804 values from the SRV records were lost.
2806 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2807 using the correct key (the original address) when searching the retry
2808 rules in order to find which one to use for generating the retry hint.
2810 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2811 adding the default one. Similarly, if it contains a Reply-To: header, the
2812 errors_reply_to option, if set, is not used.
2814 PH/19 When calculating a retry time, Exim used to measure the "time since
2815 failure" by looking at the "first failed" field in the retry record. Now
2816 it does not use this if it is later than than the arrival time of the
2817 message. Instead it uses the arrival time. This makes for better
2818 behaviour in cases where some deliveries succeed, thus re-setting the
2819 "first failed" field. An example is a quota failure for a huge message
2820 when small messages continue to be delivered. Without this change, the
2821 "time since failure" will always be short, possible causing more frequent
2822 delivery attempts for the huge message than are intended.
2823 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2825 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2826 $1, $2, $3) because the numerical variables can be reset during some
2827 expansion items (e.g. "match"), thereby losing the authentication data.
2829 PH/21 Make -bV show the size of off_t variables so that the test suite can
2830 decide whether to run tests for quotas > 2G.
2832 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2833 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2834 filecount value is greater than 2G or if a quota value is greater than 2G
2835 on a system where the size of off_t is not greater than 4, a panic error
2838 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2839 never match. The debug and -bh output now contains an explicit error
2840 message indicating a malformed IPv4 address or mask.
2842 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2843 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2844 PH/23 above applies.
2846 PH/25 Do not write to syslog when running in the test harness. The only
2847 occasion when this arises is a failure to open the main or panic logs
2848 (for which there is an explicit test).
2850 PH/26 Added the /no_tell option to "control=freeze".
2852 PH/27 If a host name lookup failed very early in a connection, for example, if
2853 the IP address matched host_lookup and the reverse lookup yielded a name
2854 that did not have a forward lookup, an error message of the form "no IP
2855 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2856 could be logged. Now it outputs the IP address instead of "NULL".
2858 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2859 allows the sender and the authenticated sender to be set when
2860 submitting a message from within Exim. Since child_open_exim() is
2861 documented for local_scan(), the new function should be too.
2863 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2864 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2865 results in an empty string is now treated as unset.
2867 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2869 PH/31 Added sender_verify_fail logging option.
2871 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2872 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2873 tidied the source and removed it altogether.
2875 PH/33 When a queue run was abandoned because the load average was too high, a
2876 log line was always written; now it is written only if the queue_run log
2877 selector is set. In addition, the log line for abandonment now contains
2878 information about the queue run such as the pid. This is always present
2879 in "start" and "stop" lines but was omitted from the "abandon" line.
2881 PH/34 Omit spaces between a header name and the colon in the error message that
2882 is given when verify = headers_syntax fails (if there are lots of them,
2883 the message gets confusing).
2885 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2886 names, as there are now some PTR records that contain slashes. This check
2887 is only to protect against broken name servers that fall over on strange
2888 characters, so the fact that it applies to all lookups doesn't matter.
2890 PH/36 Now that the new test suite is complete, we can remove some of the
2891 special code in Exim that was needed for the old test suite. For example,
2892 sorting DNS records because real resolvers return them in an arbitrary
2893 order. The new test suite's fake resolver always returns records in the
2896 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2897 bounces) except when queue_only is set, to avoid logging races between
2898 the different processes.
2900 PH/38 Panic-die if .include specifies a non-absolute path.
2902 PH/39 A tweak to the "H" retry rule from its user.
2904 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2905 a label. They prevented compilation on older perls.
2907 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2908 a warning to be raised on newish perls.
2910 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2911 on queue. Changes to match documented behaviour of showing count of
2912 messages matching specified criteria.
2914 PH/40 Changed the default ident timeout from 30s to 5s.
2916 PH/41 Added support for the use of login_cap features, on those BSD systems
2917 that have them, for controlling the resources used by pipe deliveries.
2919 PH/42 The content-scanning code uses fopen() to create files in which to put
2920 message data. Previously it was not paying any attention to the mode of
2921 the files. Exim runs with umask(0) because the rest of the code creates
2922 files with open(), and sets the required mode explicitly. Thus, these
2923 files were ending up world-writeable. This was not a big issue, because,
2924 being within the spool directory, they were not world-accessible. I have
2925 created a function called modefopen, which takes an additional mode
2926 argument. It sets umask(777), creates the file, chmods it to the required
2927 mode, then resets the umask. All the relevant calls to fopen() in the
2928 content scanning code have been changed to use this function.
2930 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2931 to 24 hours. This avoids potential overflow problems when processing G
2932 and H retry rules. I suspect nobody ever tinkers with this value.
2934 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2936 PH/45 When the plaintext authenticator is running as a client, the server's
2937 challenges are checked to ensure they are valid base64 strings. By
2938 default, the authentication attempt is cancelled if an invalid string is
2939 received. Setting client_ignore_invalid_base64 true ignores these errors.
2940 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2941 they are received. Thus, the responses can be made to depend on the
2942 challenges. If an invalid string is ignored, an empty string is placed in
2945 PH/46 Messages that are created by the autoreply transport now contains a
2946 References: header, in accordance with RFCs 2822 and 3834.
2948 PH/47 Added authenticated_sender_force to the smtp transport.
2950 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2952 PH/49 Installed latest patch from the Sieve maintainer.
2954 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2955 was also set, the appendfile transport was unnecessarily scanning a
2956 directory of message files (e.g. for maildir delivery) to find the count
2957 of files (along with the size), even though it did not need this
2958 information. It now does the scan only if it needs to find either the
2959 size of the count of files.
2961 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2963 PH/52 Two bugs concerned with error handling when the smtp transport is
2966 (i) Exim was not creating retry information for temporary errors given
2967 for individual recipients after the DATA command when the smtp transport
2968 was used in LMTP mode. This meant that they could be retried too
2969 frequently, and not timed out correctly.
2971 (ii) Exim was setting the flag that allows error details to be returned
2972 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2973 recipients that were returned after the DATA command.
2975 PH/53 This is related to PH/52, but is more general: for any failing address,
2976 when detailed error information was permitted to be returned to the
2977 sender, but the error was temporary, then after the final timeout, only
2978 "retry timeout exceeded" was returned. Now it returns the full error as
2979 well as "retry timeout exceeded".
2981 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2982 do this, and (what is worse) MTAs that accept it.
2984 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2985 will now be deprecated.
2987 PH/56 New os.c-cygwin from the Cygwin maintainer.
2989 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2990 formats (previously only available in exim formats via -bpr, -bpru,
2991 and -bpra. Now also available in native and exiqgrep formats)
2993 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2994 with very large, slow to parse queues
2996 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2998 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3000 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3001 responses to authentication challenges, though it was showing the
3002 challenges; (ii) I've removed the CR characters from the debug output for
3005 PH/58 Allow for the insertion of a newline as well as a space when a string
3006 is turned into more than one encoded-word during RFC 2047 encoding. The
3007 Sieve code now uses this.
3009 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3010 data_4xx, lost_connection, tls_required.
3012 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3013 message was being sent as an SMTP response.
3015 PH/61 Add -l and -k options to exicyclog.
3017 PH/62 When verifying, if an address was redirected to one new address, so that
3018 verification continued, and the new address failed or deferred after
3019 having set something in $address_data, the value of $address_data was not
3020 passed back to the ACL. This was different to the case when no
3021 redirection occurred. The value is now passed back in both cases.
3023 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3024 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3025 use login_cap.h, so on its own it isn't the distinguishing feature. The
3026 new name refers directly to the setclassresources() function.
3028 PH/65 Added configuration files for NetBSD3.
3030 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3032 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3033 is preferred over IPv4.
3035 PH/68 The bounce_return_message and bounce_return_body options were not being
3036 honoured for bounces generated during the reception of non-SMTP messages.
3037 In particular, this applied to messages rejected by the ACL. This bug has
3038 been fixed. However, if bounce_return_message is true and bounce_return_
3039 body is false, the headers that are returned for a non-SMTP message
3040 include only those that have been read before the error was detected.
3041 (In the case of an ACL rejection, they have all been read.)
3043 PH/69 The HTML version of the specification is now built in a directory called
3044 spec_html instead of spec.html, because the latter looks like a path with
3045 a MIME-type, and this confuses some software.
3047 PH/70 Catch two compiler warnings in sieve.c.
3049 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3050 function verify_get_ident() calls ip_connect() to connect a socket, but
3051 if the "connect()" function timed out, ip_connect() used to close the
3052 socket. However, verify_get_ident() also closes the socket later, and in
3053 between Exim writes to the log, which may get opened at this point. When
3054 the socket was closed in ip_connect(), the log could get the same file
3055 descriptor number as the socket. This naturally causes chaos. The fix is
3056 not to close the socket in ip_connect(); the socket should be closed by
3057 the function that creates it. There was only one place in the code where
3058 this was missing, in the iplookup router, which I don't think anybody now
3059 uses, but I've fixed it anyway.
3061 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3062 well as to direct DNS lookups. Otherwise the handling of names in host
3063 lists is inconsistent and therefore confusing.
3069 PH/01 Two changes to the default runtime configuration:
3071 (1) Move the checks for relay_from_hosts and authenticated clients from
3072 after to before the (commented out) DNS black list checks.
3074 (2) Add control=submission to the relay_from_hosts and authenticated
3075 clients checks, on the grounds that messages accepted by these
3076 statements are most likely to be submissions.
3078 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3080 (1) Generate an error if the third argument for the ${prvs expansion is
3083 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3086 (3) Reset the variables that are obtained from the first argument of
3087 ${prvscheck and used in the second argument before leaving the code,
3088 because their memory is reclaimed, so using them afterwards may do
3091 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3092 one (it's much easier than Tom thought :-).
3094 (5) Because of (4), we can now allow for the use of $prvscheck_result
3095 inside the third argument.
3097 PH/03 For some reason, the default setting of PATH when running a command from
3098 a pipe transport was just "/usr/bin". I have changed it to
3101 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3102 anything to be listed in the output from -bV.
3104 PH/05 When a filter generated an autoreply, the entire To: header line was
3105 quoted in the delivery log line, like this:
3107 => >A.N.Other <ano@some.domain> <original@ddress> ...
3109 This has been changed so that it extracts the operative address. There
3110 may be more than one such address. If so, they are comma-separated, like
3113 => >ano@some.domain,ona@other.domain <original@ddress> ...
3115 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3116 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3117 not being looked up in the rDNS to get a host name, Exim was showing the
3118 IP address twice in Received: lines, even though the IP addresses were
3119 identical. For example:
3121 Received: from [1.2.3.4] (helo=[1.2.3.4])
3123 However, if the real host name was known, it was omitting the HELO data
3124 if it matched the actual IP address. This has been tidied up so that it
3125 doesn't show the same IP address twice.
3127 PH/07 When both +timestamp and +memory debugging was on, the value given by
3128 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3129 called by the debug printing, thereby overwriting the timestamp buffer.
3130 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3132 PH/08 When the original message was included in an autoreply transport, it
3133 always said "this is a copy of the message, including all the headers",
3134 even if body_only or headers_only was set. It now gives an appropriate
3137 PH/09 Applied a patch from the Sieve maintainer which:
3139 o fixes some comments
3140 o adds the (disabled) notify extension core
3141 o adds some debug output for the result of if/elsif tests
3142 o points to the current vacation draft in the documentation
3143 and documents the missing references header update
3147 o fixes a bug in processing the envelope test (when testing
3148 multiple envelope elements, the last element determinted the
3151 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3152 Electronic Mail") by including:
3154 Auto-submitted: auto-generated
3156 in the messages that it generates (bounce messages and others, such as
3157 warnings). In the case of bounce messages for non-SMTP mesages, there was
3158 also a typo: it was using "Auto_submitted" (underscore instead of
3159 hyphen). Since every message generated by Exim is necessarily in response
3160 to another message, thes have all been changed to:
3162 Auto-Submitted: auto-replied
3164 in accordance with these statements in the RFC:
3166 The auto-replied keyword:
3168 - SHOULD be used on messages sent in direct response to another
3169 message by an automatic process,
3171 - MUST NOT be used on manually-generated messages,
3173 - MAY be used on Delivery Status Notifications (DSNs) and Message
3174 Disposition Notifications (MDNs),
3176 - MUST NOT be used on messages generated by automatic or periodic
3177 processes, except for messages which are automatic responses to
3180 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3181 to the default Received: header definition.
3183 PH/12 Added log selector acl_warn_skipped (default on).
3185 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3186 variables because (a) they are in the wrong storage pool and (b) even if
3187 they were copied, it wouldn't work properly because of the caching.
3189 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3190 checking when decoding. Apparently there are clients that generate
3191 overlong encoded strings. Why am I not surprised?
3193 PH/15 If the first argument of "${if match_address" was not empty, but did not
3194 contain an "@" character, Exim crashed. Now it writes a panic log message
3195 and treats the condition as false.
3197 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3199 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3200 extension "envelope-auth". The code is finished and in agreement with
3201 other implementations, but there is no documentation so far and in fact,
3202 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3203 not changing the active code.
3205 Print executed "if" and "elsif" statements when debugging is used. This
3206 helps a great deal to understand what a filter does.
3208 Document more things not specified clearly in RFC3028. I had all this
3209 sorted out, when out of a sudden new issues came to my mind. Oops."
3211 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3214 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3215 canonical form (as documented). However, after a host name lookup from
3216 the IP address, check_host() was doing a simple string comparison with
3217 addresses acquired from the DNS when checking that the found name did
3218 have the original IP as one of its addresses. Since any found IPv6
3219 addresses are likely to be in abbreviated form, the comparison could
3220 fail. Luckily, there already exists a function for doing the comparison
3221 by converting both addresses to binary, so now that is used instead of
3222 the text comparison.
3224 PH/20 There was another similar case to PH/19, when a complete host name was
3225 given in a host list; looking up its IP address could give an abbreviated
3226 form, whereas the current host's name might or might not be abbreviated.
3227 The same fix has been applied.
3233 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3234 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3237 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3238 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3240 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3242 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3243 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3244 Also, for "Auto-Submitted", treat anything other than "no" as
3245 non-personal, in accordance with RFC 3834. (Previously it treated
3246 anything starting "auto-" as non-personal.)
3248 TF/01 The control=submission/name=... option had a problem with syntax
3249 errors if the name included a slash character. The /name= option
3250 now slurps the rest of the string, so it can include any characters
3251 but it must come last in the list of options (after /sender_retain
3254 PH/05 Some modifications to the interface to the fake nameserver for the new
3262 TK/01 Added the "success_on_redirect" address verification option. See
3263 NewStuff for rationale and an example.
3265 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3267 PH/02 Patch to exigrep to allow it to work on syslog lines.
3269 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3270 fread() to skip over the body file's header line, because in Cygwin the
3271 header line is locked and is inaccessible.
3273 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3274 co-exist for some time) to make it clear that it is the Exim ID that is
3275 referenced, not the Message-ID: header line.
3277 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3278 string_format() function, because snprintf() does not exist on all
3281 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3282 ${stat: expansion item.
3284 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3285 protocol synchronization error", to keep the pedants happy.
3287 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3288 well as for IRIX systems, when gcc is being used. See the host.c source
3291 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3293 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3296 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3297 transport and to the smtp transport in LMTP mode.
3299 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3301 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3302 run a filter in a subprocess. This could lead to confusion in subsequent
3303 lookups in the parent process. There should also be a search_tidyup() at
3304 the end of the subprocess.
3306 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3307 only if the host matched helo_try_verify_hosts, which caused the
3308 verification to occur when the EHLO/HELO command was issued. The ACL just
3309 tested the remembered result. Now, if a previous verification attempt has
3310 not happened, "verify = helo" does it there and then.
3312 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3314 TK/03 Fix log output including CR from clamd.
3316 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3317 exist provoked a memory error which could cause a segfault.
3319 PH/15 Installed PCRE 6.2
3321 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3323 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3324 of the problem. Specifically, suggested +O2 rather than +O1 for the
3327 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3329 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3330 record, the port was not correctly logged when the outgoing_port log
3331 selector was set (it logged the transort's default port).
3333 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3334 fallback_hosts, and "hosts" in the smtp transport.
3336 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3337 host errors such as "Connection refused".
3339 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3340 authentication with radiusclient 0.4.9:
3342 - Error returned from rc_read_config was caught wrongly
3343 - Username/password not passed on to radius server due to wrong length.
3345 The presumption is that some radiusclient API changes for 4.51/PH/17
3346 were not taken care of correctly. The code is still untested by me (my
3347 Linux distribution still has 0.3.2 of radiusclient), but it was
3348 contributed by a Radius user.
3350 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3351 expanding the "port" option of the smtp transport.
3353 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3354 while reading a MIME header. Thanks to Tom Hughes for a patch.
3356 PH/24 Include config.h inside local_scan.h so that configuration settings are
3359 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3360 that in an ACL for RCPT (for example), you can examine exactly what was
3363 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3364 commands, but it was not correctly comparing the address with the actual
3365 client host address. Thus, it would show the EHLO address in Received:
3366 header lines when this was not necessary.
3368 PH/27 Added the % operator to ${eval:}.
3370 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3371 it should be ignoring failures (because with -C, for example, it has lost
3372 privilege). It wasn't ignoring creation failures other than "already
3375 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3378 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3379 because an input line was too long, either on its own, or by virtue of
3380 too many continuations, the temporary file was not being removed, and the
3381 return code was incorrect.
3383 PH/31 Missing "BOOL" in function definition in filtertest.c.
3385 PH/32 Applied Sieve patches from the maintainer.
3387 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3389 PH/33 Added "verify = not_blind".
3391 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3392 Local/Makefile (with some defaults set). These are used in built scripts
3393 such as exicyclog, but they have never been used in the exim_install
3394 script (though there are many overriding facilities there). I have
3395 arranged that the exim_install script now takes note of these two
3398 PH/35 Installed configuration files for Dragonfly.
3400 PH/36 When a locally submitted message by a trusted user did not contain a
3401 From: header, and the sender address was obtained from -f or from an SMTP
3402 MAIL command, and the trusted user did not use -F to supply a sender
3403 name, $originator_name was incorrectly used when constructing a From:
3404 header. Furthermore, $originator_name was used for submission mode
3405 messages from external hosts without From: headers in a similar way,
3406 which is clearly wrong.
3408 PH/37 Added control=suppress_local_fixups.
3410 PH/38 When log_selector = +received_sender was set, and the addition of the
3411 sender made the log line's construction buffer exactly full, or one byte
3412 less than full, an overflow happened when the terminating "\n" was
3415 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3416 when the result of a list match is failure because a DNS lookup failed.
3418 PH/40 RM_COMMAND is now used in the building process.
3420 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3421 the "build-* directories that it finds.
3423 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3424 address in a domain literal was a prefix of an interface address.
3426 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3427 when verifying a sender address, unless rewrite_headers is false.
3429 PH/44 Wrote a long comment about why errors_to addresses are verified as
3430 recipients, not senders.
3432 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3433 the ratelimit ACL was added.
3435 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3437 PH/46 Added extra information about PostgreSQL errors to the error string.
3439 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3440 avoiding the need to install special zones in a real server. This is
3441 backwards compatible; if it can't find the fake resolver, it drops back.
3442 Thus, both old and new test suites can be run.
3444 TF/02 Added util/ratelimit.pl
3446 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3447 clock is set back in time.
3449 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3450 Candler <B.Candler@pobox.com>.
3452 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3453 OK for addresses that are the result of redirections.
3455 PH/48 A number of further additions for the benefit of the new test suite,
3456 including a fake gethostbyname() that interfaces to the fake DNS resolver
3459 TF/06 The fix for widen_domains has also been applied to qualify_single and
3460 search_parents which are the other dnslookup options that can cause
3463 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3466 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3468 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3469 service" if it encountered an MX record with an empty target hostname.
3470 The message is now "an MX or SRV record indicated no SMTP service".
3472 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3473 if the DNS of the sending site is misconfigured. This is quite a
3474 common situation. This change restores the behaviour of treating a
3475 helo verification defer as a failure.
3477 PH/51 If self=fail was set on a router, the bounce message did not include the
3478 actual error message.
3484 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3486 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3487 command itself ended in error, the underlying message about the transport
3488 filter timeout was being overwritten with the pipe command error. Now the
3489 underlying error message should be appended to the second error message.
3491 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3493 PH/02 Reduce the amount of output that "make" produces by default. Full output
3494 can still be requested.
3496 PH/03 The warning log line about a condition test deferring for a "warn" verb
3497 was being output only once per connection, rather than after each
3498 occurrence (because it was using the same function as for successful
3499 "warn" verbs). This seems wrong, so I have changed it.
3501 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3502 it should not have, which might have caused a crash in the right
3503 circumstances, but probably never did.
3505 PH/04 Installed a modified version of Tony Finch's patch to make submission
3506 mode fix the return path as well as the Sender: header line, and to
3507 add a /name= option so that you can make the user's friendly name appear
3510 TF/03 Added the control = fakedefer ACL modifier.
3512 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3513 Mark Lowes for thorough testing.
3515 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3517 TK/03 Merged latest SRS patch from Miles Wilton.
3519 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3520 with the definition in sysexits.h (which is #included earlier).
3521 Fortunately, Exim does not actually use EX_OK. The code used to try to
3522 preserve the sysexits.h value, by assumimg that macro definitions were
3523 scanned for macro replacements. I have been disabused of this notion,
3524 so now the code just undefines EX_OK before #including unistd.h.
3526 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3527 in the smtp transport. When a block could not be written in a single
3528 write() function, the timeout was being re-applied to each part-write.
3529 This seems wrong - if the receiver was accepting one byte at a time it
3530 would take for ever. The timeout is now adjusted when this happens. It
3531 doesn't have to be particularly precise.
3533 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3534 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3536 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3537 without a domain if the check to <postmaster@domain> fails.
3539 SC/01 Eximstats: added -xls and the ability to specify output files
3540 (patch written by Frank Heydlauf).
3542 SC/02 Eximstats: use FileHandles for outputing results.
3544 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3546 SC/04 Eximstats: fixed display of large numbers with -nvr option
3548 SC/05 Eximstats: fixed merging of reports with empty tables.
3550 SC/06 Eximstats: added the -include_original_destination flag
3552 SC/07 Eximstats: removed tabs and trailing whitespace.
3554 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3556 TK/06 MBOX spool code: Add real "From " MBOX separator line
3557 so the .eml file is really in mbox format (even though
3558 most programs do not really care). Patch from Alex Miller.
3560 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3561 The latter is generated from $received_to and is only set if the
3562 message has one envelope recipient. SA can use these headers,
3563 obviously out-of-the-box. Patch from Alex Miller.
3565 PH/08 The ${def test on a variable was returning false if the variable's
3566 value was "0", contrary to what the specification has always said!
3567 The result should be true unless the variable is empty.
3569 PH/09 The syntax error of a character other than { following "${if
3570 def:variable_name" (after optional whitespace) was not being diagnosed.
3571 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3572 accidental colon was present, for example, could give incorrect results.
3574 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3575 result is used (not including appendfile, where other changes are about
3578 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3579 This involved changing a lot of size variables from int to off_t. It
3580 should work with maildirs and everything.
3582 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3583 spamd dying while we are connected to it.
3585 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3588 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3589 function for BDB 4.3.
3591 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3593 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3594 the sources for PCRE have been re-arranged and more files are now
3597 PH/15 The code I had for printing potentially long long variables in PH/11
3598 above was not the best (it lost precision). The length of off_t variables
3599 is now inspected at build time, and an appropriate printing format (%ld
3600 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3601 to be "long long int" or "long int". This is needed for the internal
3602 formatting function string_vformat().
3604 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3605 the configuration file to be ":syslog", then the script "guesses" where
3606 the logs files are, rather than using the compiled in default. In our
3607 case the guess is not the same as the compiled default, so the script
3608 suddenly stopped working when I started to use syslog. The patch checks
3609 to see if log_file_path is "". If so, it attempts to read it from exim
3610 with no configuration file to get the compiled in version, before it
3611 falls back to the previous guessing code."
3613 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3614 implementing BATV in an Exim configuration. See NewStuff for the gory
3617 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3618 Makefile that are specific to HP-UX.
3620 PH/18 If the "use_postmaster" option was set for a recipient callout together
3621 with the "random" option, the postmaster address was used as the MAIL
3622 FROM address for the random test, but not for the subsequent recipient
3623 test. It is now used for both.
3625 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3626 patch removes a few documentation additions to RFC 3028, because the
3627 latest draft now contains them. It adds the new en;ascii-case comparator
3628 and a new error check for 8bit text in MIME parts. Comparator and
3629 require names are now matched exactly. I enabled the subaddress
3630 extension, but it is not well tested yet (read: it works for me)."
3632 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3633 rework some of the code of TK/09 above to avoid the hardwired use of
3634 "%lld" and "long long". Replaced the call to snprintf() with a call to
3637 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3638 records point to non-existent hosts", "retry timeout exceeded", and
3639 "retry time not reached for any host after a long failure period".
3641 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3642 experimental DomainKeys support:
3644 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3645 (2) On an error such as an illegally used "control", the wrong name for
3646 the control was given.
3648 These problems did NOT occur unless DomainKeys support was compiled.
3650 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3652 PH/24 Added ${if match_ip condition.
3654 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3655 databases so that it will be absolutely obvious if a crash occurs in the
3656 DB library. This is a regular occurrence (often caused by mis-matched
3659 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3660 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3661 picked up on a user's system that detects such things. There doesn't seem
3662 to be a gcc warning option for this - only an attribute that has to be
3663 put on the function's prototype. It seems that in Fedora Core 4 they have
3664 set this on a number of new functions. No doubt there will be more in due
3667 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3668 specify a transport. However, if an address that was verified by such a
3669 router was the subject of a callout, Exim crashed because it tried to
3670 read the rcpt_include_affixes from the non-existent transport. Now it
3671 just assumes that the setting of that option is false. This bug was
3672 introduced by 4.51/PH/31.
3674 PH/28 Changed -d+all to exclude +memory, because that information is very
3675 rarely of interest, but it makes the output a lot bigger. People tend to
3676 do -d+all out of habit.
3678 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3679 code in os-type was giving problems when libc.so lives in lib64, like on
3682 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3683 aren't the modern standard, and it seems that some systems' include files
3684 don't always have them. Exim was already checking for some of the newer
3685 ones like T_AAAA, and defining it itself. I've added checks for all the
3686 record types that Exim uses.
3688 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3689 not automatically generating a new one, as it is supposed to. This
3690 prevented TLS from working. If the file did exist, but contained invalid
3691 data, a new version was generated, as expected. It was only the case of a
3692 non-existent file that was broken.
3694 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3695 with a change in libdomainkeys > 0.64.
3697 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3698 from DNS. If the selector record carries the flag, it now has
3699 precedence over the domain-wide flag.
3701 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3703 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3704 the use of a port name that isn't defined in /etc/services) occurred, the
3705 message was deferred as in a normal delivery, and thus remained on the
3706 spool, instead of being failed because of the mua_wrapper setting. This
3707 is now fixed, and I tidied up some of the mua_wrapper messages at the
3710 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3711 the messages in a hash of arrays rather than using individual hashes.
3712 This is a bit cleaner and results in dramatic memory savings, albeit
3713 at a slight CPU cost.
3715 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3716 as requested by Marc Sherman.
3718 SC/10 Eximstats: added histograms for user specified patterns as requested
3721 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3723 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3724 fopen() in the content-scanning modules that did not already have it.
3730 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3731 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3733 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3735 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3737 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3738 to test Sieve filters that use "vacation".
3740 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3741 that changes the way the GnuTLS parameters are stored in the cache file.
3742 The new format can be generated externally. For backward compatibility,
3743 if the data in the cache doesn't make sense, Exim assumes it has read an
3744 old-format file, and it generates new data and writes a new file. This
3745 means that you can't go back to an older release without removing the
3748 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3749 work if there are any delivery delays because "one_time" forces the
3750 parent to be marked "delivered", so its unseen clone is never tried
3751 again. For this reason, Exim now forbids the simultaneous setting of
3754 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3755 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3756 entry if you want to know the details. The fix, however, neglected to
3757 consider the case when local delivery batching is involved. The test for
3758 "previously delivered" was not happening when checking to see if an
3759 address could be batched with a previous (undelivered) one; under
3760 certain circumstances this could lead to multiple deliveries to the same
3763 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3764 in its include files, and this causes problems building Exim.
3766 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3767 header_syntax) but Exim was just ignoring anything given after a slash.
3768 In particular, this caused confusion with an attempt to use "verify =
3769 reverse_host_lookup/defer_ok". An error is now given when options are
3770 supplied for verify items that do not have them. (Maybe reverse_host_
3771 lookup should have a defer_ok option, but that's a different point.)
3773 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3774 defined by RFC 821) to 2048, because there were problems with some AUTH
3775 commands, and RFC 1869 says the size should be increased for extended
3776 SMTP commands that take arguments.
3778 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3781 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3782 "unknown" error; now it says that the functionality isn't in the binary.
3784 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3785 an address' error message when a string expansion fails (syntax or
3786 whatever). Otherwise the password may appear in the log. Following change
3787 PH/42 below, there is no longer a chance of it appearing in a bounce
3790 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3792 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3793 of its addresses was ever tried. (Bugzilla bug #2.)
3795 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3796 the result incorrectly in the debug output. (It correctly added a newline
3797 to what was transported.)
3799 TF/01 Added $received_time.
3801 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3802 commented out examples of how to interface to a virus scanner and to
3803 SpamAssassin. Also added commented examples of av_scanner and
3804 spamd_address settings.
3806 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3807 and controls are allowed in which ACLs. There were a couple of minor
3808 errors. Some of the entries in the conditions table (which is a table of
3809 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3810 negation of where the condition IS allowed.
3812 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3814 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3815 header file does not have a version number, so I've had to invent a new
3816 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3817 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3818 radiusclient), but it was contributed by a Radius user.
3820 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3821 files or directories, for OpenSSL.
3823 PH/19 When an Exim process that is running as root has to create an Exim log
3824 file, it does so in a subprocess that runs as exim:exim so as to get the
3825 ownership right at creation (otherwise, other Exim processes might see
3826 the file with the wrong ownership). There was no test for failure of this
3827 fork() call, which would lead to the process getting stuck as it waited
3828 for a non-existent subprocess. Forks do occasionally fail when resources
3829 run out. I reviewed all the other calls to fork(); they all seem to check
3832 PH/20 When checking for unexpected SMTP input at connect time (before writing
3833 the banner), Exim was not dealing correctly with a non-positive return
3834 from the read() function. If the client had disconnected by this time,
3835 the result was a log entry for a synchronization error with an empty
3836 string after "input=" when read() returned zero. If read() returned -1
3837 (an event I could not check), uninitialized data bytes were printed.
3838 There were reports of junk text (parts of files, etc) appearing after
3841 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3843 PH/22 Added support for macro redefinition, and (re)definition in between
3844 driver and ACL definitions.
3846 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3847 forgetting to use the resulting value; it was using the unexpanded value.
3849 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3850 hadn't been configured. The fix is from Juergen Kreileder, who
3851 understands it better than I do:
3853 "Here's what I see happening with three configured cyrus_sasl
3854 authenticators configured (plain, login, cram-md5):
3856 On startup auth_cyrus_sasl_init() gets called for each of these.
3857 This means three calls to sasl_listmech() without a specified mech_list.
3858 => SASL tests which mechs of all available mechs actually work
3859 => three warnings about OTP not working
3860 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3862 With the patch, sasl_listmech() also gets called three times. But now
3863 SASL's mech_list option is set to the server_mech specified in the the
3864 authenticator. Or in other words, the answer from sasl_listmech()
3865 gets limited to just the mech you're testing for (which is different
3867 => the return list contains just 'plain' or 'login', 'cram-md5' or
3868 nothing depending on the value of ob->server_mech.
3870 I've just tested the patch: Authentication still works fine,
3871 unavailable mechs specified in the exim configuration are still
3872 caught, and the auth.log warnings about OTP are gone."
3874 PH/25 When debugging is enabled, the contents of the command line are added
3875 to the debugging output, even when log_selector=+arguments is not
3878 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3879 answer is "GNU", and only if the return is "GNU/something" is the answer
3882 PH/27 $acl_verify_message is now set immediately after the failure of a
3883 verification in an ACL, and so is available in subsequent modifiers. In
3884 particular, the message can be preserved by coding like this:
3886 warn !verify = sender
3887 set acl_m0 = $acl_verify_message
3889 Previously, $acl_verify_message was set only while expanding "message"
3890 and "log_message" when a very denied access.
3892 PH/28 Modified OS/os.c-Linux with
3894 -#ifndef OS_LOAD_AVERAGE
3895 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3897 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3898 nomenclature these days.)
3900 PH/29 Installed patch from the Sieve maintainer that adds the options
3901 sieve_useraddress and sieve_subaddress to the redirect router.
3903 PH/30 In these circumstances:
3904 . Two addresses routed to the same list of hosts;
3905 . First host does not offer TLS;
3906 . First host accepts first address;
3907 . First host gives temporary error to second address;
3908 . Second host offers TLS and a TLS session is established;
3909 . Second host accepts second address.
3910 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3911 and peerdn, if requested) that were in fact used only for the second
3914 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3915 attention to any local part prefix or suffix that was matched by the
3916 router that accepted the address. It now behaves in the same way as it
3917 does for delivery: the affixes are removed from the local part unless
3918 rcpt_include_affixes is set on the transport.
3920 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3921 timeout during the DATA phase of an incoming message.
3923 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3924 applied a patch sanctioned by the Sieve maintainer.
3926 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3927 the uid or gid is negative. A case of a negative gid caused this to be
3928 noticed. The fix allows for either to be negative.
3930 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3931 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3934 PH/36 The change PH/12 above was broken. Fixed it.
3936 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3937 the grounds that routing the same address twice would always produce the
3938 same answer. This might have been true once, but it is certainly no
3939 longer true now. Routing a child address may depend on the previous
3940 routing that produced that child. Some complicated redirection strategies
3941 went wrong when messages had multiple recipients, and made Exim's
3942 behaviour dependent on the order in which the addresses were given.
3944 I have moved the duplicate checking until after the routing is complete.
3945 Exim scans the addresses that are assigned to local and remote
3946 transports, and removes any duplicates. This means that more work will be
3947 done, as duplicates will always all be routed, but duplicates are
3948 presumably rare, so I don't expect this is of any significance.
3950 For deliveries to pipes, files, and autoreplies, the duplicate checking
3951 still happens during the routing process, since they are not going to be
3954 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3955 It corrects a timeout issue with spamd. This is Ian's comment: "The
3956 background is that sometimes spamd either never reads data from a
3957 connection it has accepted, or it never writes response data. The exiscan
3958 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3959 blindly assumes that writes won't block so it may never time out."
3961 PH/39 Allow G after quota size as well as K and M.
3963 PH/40 The value set for $authenticated_id in an authenticator may not contain
3964 binary zeroes or newlines because the value is written to log lines and
3965 to spool files. There was no check on this. Now the value is run through
3966 the string_printing() function so that such characters are converted to
3967 printable escape sequences.
3969 PH/41 $message_linecount is a new variable that contains the total number of
3970 lines in the message. Compare $body_linecount, which is the count for the
3973 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3974 bounce and delay warning messages, except in certain special cases, which
3977 (a) An SMTP error message from a remote host;
3978 (b) A message specified in a :fail: redirection;
3979 (c) A message specified in a "fail" command in a system filter;
3980 (d) A message specified in a FAIL return from the queryprogram router;
3981 (e) A message specified by the cannot_route_message router option.
3983 In these cases only, Exim does include the error details in bounce and
3984 warning messages. There are also a few cases where bland messages such
3985 as "unrouteable address" or "local delivery error" are given.
3987 PH/43 $value is now also set for the "else" part of a ${run expansion.
3989 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3990 being worked on, but at least Exim now implements the latest version to
3993 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3994 process to complete was treated as a delivery failure, a timeout while
3995 writing the message to the pipe was logged, but erroneously treated as a
3996 successful delivery. Such timeouts include transport filter timeouts. For
3997 consistency with the overall process timeout, these timeouts are now
3998 treated as errors, giving rise to delivery failures by default. However,
3999 there is now a new Boolean option for the pipe transport called
4000 timeout_defer, which, if set TRUE, converts the failures into defers for
4001 both kinds of timeout. A transport filter timeout is now identified in
4004 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4005 systems where "make" and "gmake" are different, calling "gmake" at top
4006 level broke things. I've arranged for the value of $(MAKE) to be passed
4007 from the Makefile to this script so that it can call the same version of
4011 A note about Exim versions 4.44 and 4.50
4012 ----------------------------------------
4014 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4015 changes of various kinds. As a consequence, a big documentation update was
4016 needed. This delayed the release for rather longer than seemed good, especially
4017 in the light of a couple of (minor) security issues. Therefore, the changes
4018 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4019 release. So 4.44 and 4.50 are in effect two different branches that both start
4022 I have left the 4.50 change log unchanged; it contains all the changes since
4023 4.43. The change log for 4.44 is below; many of its items are identical to
4024 those for 4.50. This seems to be the most sensible way to preserve the
4025 historical information.
4031 1. Minor wording change to the doc/README.SIEVE file.
4033 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4034 computation of the current number of files was incorrect.
4036 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4037 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4040 4. Give more explanation in the error message when the command for a transport
4041 filter fails to execute.
4043 5. There are several places where Exim runs a non-Exim command in a
4044 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4045 was being done only for the command run by the queryprogram router. It is
4046 now done for all such subprocesses. The other cases are: ${run, transport
4047 filters, and the commands run by the lmtp and pipe transports.
4049 6. Added CONFIGURE_GROUP build-time option.
4051 7. Some older OS have a limit of 256 on the maximum number of file
4052 descriptors. Exim was using setrlimit() to set 1000 as a large value
4053 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4054 systems. I've change it so that if it can't get 1000, it tries for 256.
4056 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4057 was an oversight, and furthermore, ever since the addition of extra
4058 controls (e.g. 4.43/32), the checks on when to allow different forms of
4059 "control" were broken. There should now be diagnostics for all cases when a
4060 control that does not make sense is encountered.
4062 9. Added the /retain_sender option to "control=submission".
4064 10. $recipients is now available in the predata ACL (oversight).
4066 11. Tidy the search cache before the fork to do a delivery from a message
4067 received from the command line. Otherwise the child will trigger a lookup
4068 failure and thereby defer the delivery if it tries to use (for example) a
4069 cached ldap connection that the parent has called unbind on.
4071 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4072 of $address_data from the recipient verification was clobbered by the
4073 sender verification.
4075 13. The value of address_data from a sender verification is now available in
4076 $sender_address_data in subsequent conditions in the ACL statement.
4078 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4080 15. Added a new option "connect=<time>" to callout options, to set a different
4083 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4084 was its contents. (It was OK if the option was not defined at all.)
4086 17. A "Completed" log line is now written for messages that are removed from
4087 the spool by the -Mrm option.
4089 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4090 information about exactly what failed.
4092 19. Added -dd to debug only the daemon process.
4094 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4095 handles timeouts, both on the server side and network timeouts. Renamed the
4096 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4098 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4099 It is now set to "smtps".
4101 22. $host_address is now set to the target address during the checking of
4102 ignore_target_hosts.
4104 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4105 being passed; this would have caused $sender_host_name to have been used if
4106 matching the list had actually called for a host name (not very likely,
4107 since this list is usually IP addresses). A host name is now passed as
4110 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4111 code that specifies a non-restarting handler (typically sigaction() in
4112 modern systems) in an attempt to fix a rare and obscure crash bug.
4114 25. Narrowed the window for a race in the daemon that could cause it to ignore
4115 SIGCHLD signals. This is not a major problem, because they are used only to
4116 wake it up if nothing else does.
4118 26. A malformed maildirsize file could cause Exim to calculate negative values
4119 for the mailbox size or file count. Odd effects could occur as a result.
4120 The maildirsize information is now recalculated if the size or filecount
4123 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4124 support for a long time. Removed HAVE_SYS_VFS_H.
4126 28. Installed the latest version of exipick from John Jetmore.
4128 29. In an address list, if the pattern was not a regular expression, an empty
4129 subject address (from a bounce message) matched only if the pattern was an
4130 empty string. Non-empty patterns were not even tested. This was the wrong
4131 because it is perfectly reasonable to use an empty address as part of a
4132 database query. An empty address is now tested by patterns that are
4133 lookups. However, all the other forms of pattern expect the subject to
4134 contain a local part and a domain, and therefore, for them, an empty
4135 address still always fails if the pattern is not itself empty.
4137 30. Exim went into a mad DNS loop when attempting to do a callout where the
4138 host was specified on an smtp transport, and looking it up yielded more
4139 than one IP address.
4141 31. Re-factored the code for checking spool and log partition space into a
4142 function that finds that data and another that does the check. The former
4143 is then used to implement four new variables: $spool_space, $log_space,
4144 $spool_inodes, and $log_inodes.
4146 32. The RFC2047 encoding function was originally intended for short strings
4147 such as real names; it was not keeping to the 75-character limit for
4148 encoded words that the RFC imposes. It now respects the limit, and
4149 generates multiple encoded words if necessary. To be on the safe side, I
4150 have increased the buffer size for the ${rfc2047: expansion operator from
4153 33. It is now permitted to omit both strings after an "if" condition; if the
4154 condition is true, the result is "true". As before, when the second string
4155 is omitted, a false condition yields an empty string. This makes it less
4156 cumbersome to write custom ACL and router conditions.
4158 34. Failure to deliver a bounce message always caused it to be frozen, even if
4159 there was an errors_to setting on the router. The errors_to setting is now
4162 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4163 canonical form (fully expanded) before being placed in
4164 $sender_host_address.
4166 36. The table in the code that translates DNS record types into text (T_A to
4167 "A" for instance) was missing entries for NS and CNAME. It is just possible
4168 that this could have caused confusion if both these types were looked up
4169 for the same domain, because the text type is used as part of Exim's
4170 per-process caching. But the chance of anyone hitting this buglet seems
4173 37. The dnsdb lookup has been extended in a number of ways.
4175 (1) There is a new type, "zns", which walks up the domain tree until it
4176 finds some nameserver records. It should be used with care.
4178 (2) There is a new type, "mxh", which is like "mx" except that it returns
4179 just the host names, not the priorities.
4181 (3) It is now possible to give a list of domains (or IP addresses) to be
4182 looked up. The behaviour when one of the lookups defers can be
4183 controlled by a keyword.
4185 (4) It is now possible to specify the separator character for use when
4186 multiple records are returned.
4188 38. The dnslists ACL condition has been extended: it is now possible to supply
4189 a list of IP addresses and/or domains to be looked up in a particular DNS
4192 39. Added log_selector=+queue_time_overall.
4194 40. When running the queue in the test harness, wait just a tad after forking a
4195 delivery process, to get repeatability of debugging output.
4197 41. Include certificate and key file names in error message when GnuTLS fails
4198 to set them up, because the GnuTLS error message doesn't include the name
4199 of the failing file when there is a problem reading it.
4201 42. Allow both -bf and -bF in the same test run.
4203 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4205 44. The "Exiscan patch" is now merged into the mainline Exim source.
4207 45. Sometimes the final signoff response after QUIT could fail to get
4208 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4209 < 0 before doing a fflush(). This bug looks as though it goes back to the
4210 introduction of TLS in release 3.20, but "sometimes" must have been rare
4211 because the tests only now provoked it.
4213 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4214 (this can affect the format of dates).
4216 47. exim_tidydb, when checking for the continued existence of a message for
4217 which it has found a message-specific retry record, was not finding
4218 messages that were in split spool directories. Consequently, it was
4219 deleting retry records that should have stayed in existence.
4221 48. Steve fixed some bugs in eximstats.
4223 49. The SPA authentication driver was not abandoning authentication and moving
4224 on to the next authenticator when an expansion was forced to fail,
4225 contradicting the general specification for all authenticators. Instead it
4226 was generating a temporary error. It now behaves as specified.
4228 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4229 (the order specifies the preference for clients). The order is now AES256,
4230 AES128, 3DES, ARCFOUR128.
4232 51. Small patch to Sieve code - explicitly set From: when generating an
4235 52. Exim crashed if a remote delivery caused a very long error message to be
4236 recorded - for instance if somebody sent an entire SpamAssassin report back
4237 as a large number of 550 error lines. This bug was coincidentally fixed by
4238 increasing the size of one of Exim's internal buffers (big_buffer) that
4239 happened as part of the Exiscan merge. However, to be on the safe side, I
4240 have made the code more robust (and fixed the comments that describe what
4243 53. Now that there can be additional text after "Completed" in log lines (if
4244 the queue_time_overall log selector is set), a one-byte patch to exigrep
4245 was needed to allow it to recognize "Completed" as not the last thing in
4248 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4249 patch that reportedly fixes this has been added. I am not expert enough to
4250 create a test for it. This is what the patch creator wrote:
4252 "I found a little strange behaviour of ldap code when working with
4253 Windows 2003 AD Domain, where users was placed in more than one
4254 Organization Units. When I tried to give exim partial DN, the exit code
4255 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4256 But simultaneously result of request was absolutely normal ldap result,
4257 so I produce this patch..."
4259 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4260 REFERENCE, so I have modified the code to exclude the patch when that macro
4263 55. Some experimental protocols are using DNS PTR records for new purposes. The
4264 keys for these records are domain names, not reversed IP addresses. The
4265 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4266 leaves it alone. Component reversal etc. now happens only for IP addresses.
4269 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4271 57. Double the size of the debug message buffer (to 2048) so that more of very
4272 long debug lines gets shown.
4274 58. The exicyclog utility now does better if the number of log files to keep
4275 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4277 59. Two changes related to the smtp_active_hostname option:
4279 (1) $smtp_active_hostname is now available as a variable.
4280 (2) The default for smtp_banner uses $smtp_active_hostname instead
4281 of $primary_hostname.
4283 60. The host_aton() function is supposed to be passed a string that is known
4284 to be a valid IP address. However, in the case of IPv6 addresses, it was
4285 not checking this. This is a hostage to fortune. Exim now panics and dies
4286 if the condition is not met. A case was found where this could be provoked
4287 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4288 components; fortuitously, this particular loophole had already been fixed
4289 by change 4.50/55 above.
4291 If there are any other similar loopholes, the new check in host_aton()
4292 itself should stop them being exploited. The report I received stated that
4293 data on the command line could provoke the exploit when Exim was running as
4294 exim, but did not say which command line option was involved. All I could
4295 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4296 running as the user.
4299 61. There was a buffer overflow vulnerability in the SPA authentication code
4300 (which came originally from the Samba project). I have added a test to the
4301 spa_base64_to_bits() function which I hope fixes it.
4304 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4305 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4307 63. The daemon start-up calls getloadavg() while still root for those OS that
4308 need the first call to be done as root, but it missed one case: when
4309 deliver_queue_load_max is set with deliver_drop_privilege. This is
4310 necessary for the benefit of the queue runner, because there is no re-exec
4311 when deliver_drop_privilege is set.
4313 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4314 This has been fixed.
4316 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4317 was in use, was not putting the data itself into the right store pool;
4318 consequently, it could be overwritten for a subsequent message in the same
4319 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4322 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4324 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4325 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4326 the function were treating the return as a boolean value, which happened to
4327 work because 0=false and not-0=true, but is not correct code.
4329 68. The host_aton() function was not handling scoped IPv6 addresses (those
4330 with, for example, "%eth0" on the end) correctly.
4332 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4333 negated items (that is, ~something) in unsigned ints. Some compilers
4334 apparently mutter when there is no cast.
4336 70. If an address verification called from an ACL failed, and did not produce a
4337 user-specific message (i.e. there was only a "system" message), nothing was
4338 put in $acl_verify_message. In this situation, it now puts the system
4341 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4342 session; change 4.31/43 added the unwanted input to the log line - except
4343 that it did not do this in the start of session case. It now does.
4345 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4346 This is wrong and can cause the other end to generate a synchronization
4347 error if it is another Exim or anything else that does the synchronization
4348 check. A QUIT command is no longer sent after a timeout.
4350 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4351 during host lookups.
4353 74. The defer_ok option of callout verification was not working if it was used
4354 when verifying addresses in header lines, that is, for this case:
4356 verify = header_sender/callout=defer_ok
4358 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4359 those file descriptors could be used for SMTP connections. If anything
4360 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4361 could be sent to the SMTP client, causing chaos. The daemon now opens
4362 stdin, stdout, and stderr to /dev/null when it puts itself into the
4365 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4366 log by default. The user can override this with suitable Perl magic.
4368 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4369 the log message when discard triggers, was not working for the DATA ACL or
4370 for the non-SMTP ACL.
4372 78. Error message wording change in sieve.c.
4374 79. If smtp_accept_max_per_host was set, the number of connections could be
4375 restricted to fewer than expected, because the daemon was trying to set up
4376 a new connection before checking whether the processes handling previous
4377 connections had finished. The check for completed processes is now done
4378 earlier. On busy systems, this bug wouldn't be noticed because something
4379 else would have woken the daemon, and it would have reaped the completed
4382 80. If a message was submitted locally by a user whose login name contained one
4383 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4384 It caused a spool format error. I have fixed the spool reading code. A
4385 related problem was that the "from" clause in the Received: line became
4386 illegal because of the space(s). It is now covered by ${quote_local_part.
4388 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4391 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4393 83. Patch from Sieve maintainer to fix unterminated string problem in
4394 "vacation" handling.
4396 84. Some minor changes to the Linux configuration files to help with other
4397 OS variants using glibc.
4399 85. One more patch for Sieve to update vacation handling to latest spec.
4402 ----------------------------------------------------
4403 See the note above about the 4.44 and 4.50 releases.
4404 ----------------------------------------------------
4410 1. Change 4.43/35 introduced a bug that caused file counts to be
4411 incorrectly computed when quota_filecount was set in an appendfile
4414 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4415 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4418 3. Give more explanation in the error message when the command for a transport
4419 filter fails to execute.
4421 4. There are several places where Exim runs a non-Exim command in a
4422 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4423 was being done only for the command run by the queryprogram router. It is
4424 now done for all such subprocesses. The other cases are: ${run, transport
4425 filters, and the commands run by the lmtp and pipe transports.
4427 5. Some older OS have a limit of 256 on the maximum number of file
4428 descriptors. Exim was using setrlimit() to set 1000 as a large value
4429 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4430 systems. I've change it so that if it can't get 1000, it tries for 256.
4432 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4433 was an oversight, and furthermore, ever since the addition of extra
4434 controls (e.g. 4.43/32), the checks on when to allow different forms of
4435 "control" were broken. There should now be diagnostics for all cases when a
4436 control that does not make sense is encountered.
4438 7. $recipients is now available in the predata ACL (oversight).
4440 8. Tidy the search cache before the fork to do a delivery from a message
4441 received from the command line. Otherwise the child will trigger a lookup
4442 failure and thereby defer the delivery if it tries to use (for example) a
4443 cached ldap connection that the parent has called unbind on.
4445 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4446 of $address_data from the recipient verification was clobbered by the
4447 sender verification.
4449 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4450 was its contents. (It was OK if the option was not defined at all.)
4452 11. A "Completed" log line is now written for messages that are removed from
4453 the spool by the -Mrm option.
4455 12. $host_address is now set to the target address during the checking of
4456 ignore_target_hosts.
4458 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4459 being passed; this would have caused $sender_host_name to have been used if
4460 matching the list had actually called for a host name (not very likely,
4461 since this list is usually IP addresses). A host name is now passed as
4464 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4465 code that specifies a non-restarting handler (typically sigaction() in
4466 modern systems) in an attempt to fix a rare and obscure crash bug.
4468 15. Narrowed the window for a race in the daemon that could cause it to ignore
4469 SIGCHLD signals. This is not a major problem, because they are used only to
4470 wake it up if nothing else does.
4472 16. A malformed maildirsize file could cause Exim to calculate negative values
4473 for the mailbox size or file count. Odd effects could occur as a result.
4474 The maildirsize information is now recalculated if the size or filecount
4477 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4478 support for a long time. Removed HAVE_SYS_VFS_H.
4480 18. Updated exipick to current release from John Jetmore.
4482 19. Allow an empty sender to be matched against a lookup in an address list.
4483 Previously the only cases considered were a regular expression, or an
4486 20. Exim went into a mad DNS lookup loop when doing a callout where the
4487 host was specified on the transport, if the DNS lookup yielded more than
4490 21. The RFC2047 encoding function was originally intended for short strings
4491 such as real names; it was not keeping to the 75-character limit for
4492 encoded words that the RFC imposes. It now respects the limit, and
4493 generates multiple encoded words if necessary. To be on the safe side, I
4494 have increased the buffer size for the ${rfc2047: expansion operator from
4497 22. Failure to deliver a bounce message always caused it to be frozen, even if
4498 there was an errors_to setting on the router. The errors_to setting is now
4501 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4502 canonical form (fully expanded) before being placed in
4503 $sender_host_address.
4505 24. Updated eximstats to version 1.33
4507 25. Include certificate and key file names in error message when GnuTLS fails
4508 to set them up, because the GnuTLS error message doesn't include the name
4509 of the failing file when there is a problem reading it.
4511 26. Expand error message when OpenSSL has problems setting up cert/key files.
4514 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4515 (this can affect the format of dates).
4517 28. exim_tidydb, when checking for the continued existence of a message for
4518 which it has found a message-specific retry record, was not finding
4519 messages that were in split spool directories. Consequently, it was
4520 deleting retry records that should have stayed in existence.
4522 29. eximstats updated to version 1.35
4523 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4524 1.35 - bugfix such that pie charts by volume are generated correctly
4526 30. The SPA authentication driver was not abandoning authentication and moving
4527 on to the next authenticator when an expansion was forced to fail,
4528 contradicting the general specification for all authenticators. Instead it
4529 was generating a temporary error. It now behaves as specified.
4531 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4532 (the order specifies the preference for clients). The order is now AES256,
4533 AES128, 3DES, ARCFOUR128.
4535 31. Small patch to Sieve code - explicitly set From: when generating an
4538 32. Exim crashed if a remote delivery caused a very long error message to be
4539 recorded - for instance if somebody sent an entire SpamAssassin report back
4540 as a large number of 550 error lines. This bug was coincidentally fixed by
4541 increasing the size of one of Exim's internal buffers (big_buffer) that
4542 happened as part of the Exiscan merge. However, to be on the safe side, I
4543 have made the code more robust (and fixed the comments that describe what
4546 33. Some experimental protocols are using DNS PTR records for new purposes. The
4547 keys for these records are domain names, not reversed IP addresses. The
4548 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4549 leaves it alone. Component reversal etc. now happens only for IP addresses.
4552 34. The host_aton() function is supposed to be passed a string that is known
4553 to be a valid IP address. However, in the case of IPv6 addresses, it was
4554 not checking this. This is a hostage to fortune. Exim now panics and dies
4555 if the condition is not met. A case was found where this could be provoked
4556 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4557 components; fortuitously, this particular loophole had already been fixed
4558 by change 4.50/55 or 4.44/33 above.
4560 If there are any other similar loopholes, the new check in host_aton()
4561 itself should stop them being exploited. The report I received stated that
4562 data on the command line could provoke the exploit when Exim was running as
4563 exim, but did not say which command line option was involved. All I could
4564 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4565 running as the user.
4568 35. There was a buffer overflow vulnerability in the SPA authentication code
4569 (which came originally from the Samba project). I have added a test to the
4570 spa_base64_to_bits() function which I hope fixes it.
4573 36. The daemon start-up calls getloadavg() while still root for those OS that
4574 need the first call to be done as root, but it missed one case: when
4575 deliver_queue_load_max is set with deliver_drop_privilege. This is
4576 necessary for the benefit of the queue runner, because there is no re-exec
4577 when deliver_drop_privilege is set.
4579 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4580 was in use, was not putting the data itself into the right store pool;
4581 consequently, it could be overwritten for a subsequent message in the same
4582 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4585 38. Sometimes the final signoff response after QUIT could fail to get
4586 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4587 < 0 before doing a fflush(). This bug looks as though it goes back to the
4588 introduction of TLS in release 3.20, but "sometimes" must have been rare
4589 because the tests only now provoked it.
4595 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4596 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4597 Now it returns an integer. A number of calls were still expecting a T/F
4598 return. Fortuitously, in all cases, the tests worked in OK situations,
4599 which is the norm. However, things would have gone wrong on any write
4600 failures on the smtp file descriptor. This function is used when sending
4601 messages over SMTP and also when doing verify callouts.
4603 2. When Exim is called to do synchronous delivery of a locally submitted
4604 message (the -odf or -odi options), it no longer closes stderr before doing
4607 3. Implemented the mua_wrapper option.
4609 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4611 5. Implemented the functions header_remove(), header_testname(),
4612 header_add_at_position(), and receive_remove_recipient(), and exported them
4615 6. If an ACL "warn" statement specified the addition of headers, Exim already
4616 inserted X-ACL-Warn: at the start if there was no header name. However, it
4617 was not making this test for the second and subsequent header lines if
4618 there were newlines in the string. This meant that an invalid header could
4619 be inserted if Exim was badly configured.
4621 7. Allow an ACL "warn" statement to add header lines at the start or after all
4622 the Received: headers, as well as at the end.
4624 8. Added the rcpt_4xx retry error code.
4626 9. Added postmaster_mailfrom=xxx to callout verification option.
4628 10. Added mailfrom=xxxx to the callout verification option, for verify=
4631 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4632 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4634 12. Inserted some casts to stop certain compilers warning when using pointer
4635 differences as field lengths or precisions in printf-type calls (mostly
4636 affecting debugging statements).
4638 13. Added optional readline() support for -be (dynamically loaded).
4640 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4641 same clock tick as a message's arrival, so that its received time was the
4642 same as the "first fail" time on the retry record, and that message
4643 remained on the queue past the ultimate address timeout, every queue runner
4644 would try a delivery (because it was past the ultimate address timeout) but
4645 after another failure, the ultimate address timeout, which should have then
4646 bounced the address, did not kick in. This was a "< instead of <=" error;
4647 in most cases the first failure would have been in the next clock tick
4648 after the received time, and all would be well.
4650 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4651 being recognized when the domain list was tested by the match_domain
4652 condition in an expansion string.
4654 16. Added the ${str2b64: operator.
4656 17. Exim was always calling setrlimit() to set a large limit for the number of
4657 processes, without checking whether the existing limit was already
4658 adequate. (It did check for the limit on file descriptors.) Furthermore,
4659 errors from getrlimit() and setrlimit() were being ignored. Now they are
4660 logged to the main and panic logs, but Exim does carry on, to try to do its
4661 job under whatever limits there are.
4663 18. Imported PCRE 5.0.
4665 19. Trivial typo in log message " temporarily refused connection" (the leading
4668 20. If the log selector return_path_on_delivery was set and an address was
4669 redirected to /dev/null, the delivery process crashed because it assumed
4670 that a return path would always be set for a "successful" delivery. In this
4671 case, the whole delivery is bypassed as an optimization, and therefore no
4674 21. Internal re-arrangement: the function for sending a challenge and reading
4675 a response while authentication was assuming a zero-terminated challenge
4676 string. It's now changed to take a pointer and a length, to allow for
4677 binary data in such strings.
4679 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4681 23. Exim was not respecting finduser_retries when seeking the login of the
4682 uid under which it was called; it was always trying 10 times. (The default
4683 setting of finduser_retries is zero.) Also, it was sleeping after the final
4684 failure, which is pointless.
4686 24. Implemented tls_on_connect_ports.
4688 25. Implemented acl_smtp_predata.
4690 26. If the domain in control=submission is set empty, Exim assumes that the
4691 authenticated id is a complete email address when it generates From: or
4692 Sender: header lines.
4694 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4695 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4696 chown and chgrp in /bin and hostname in /usr/bin.
4698 28. Exim was keeping the "process log" file open after each use, just as it
4699 does for the main log. This opens the possibility of it remaining open for
4700 long periods when the USR1 signal hits a daemon. Occasional processlog
4701 errors were reported, that could have been caused by this. Anyway, it seems
4702 much more sensible not to leave this file open at all, so that is what now
4705 29. The long-running daemon process does not normally write to the log once it
4706 has entered its main loop, and it closes the log before doing so. This is
4707 so that log files can straightforwardly be renamed and moved. However,
4708 there are a couple of unusual error situations where the daemon does write
4709 log entries, and I had neglected to close the log afterwards.
4711 30. The text of an SMTP error response that was received during a remote
4712 delivery was being truncated at 512 bytes. This is too short for some of
4713 the long messages that one sometimes sees. I've increased the limit to
4716 31. It is now possible to make retry rules that apply only when a message has a
4717 specific sender, in particular, an empty sender.
4719 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4720 it possible to be selective about when SMTP synchronization is enforced.
4722 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4724 32. Implemented hosts_connection_nolog.
4726 33. Added an ACL for QUIT.
4728 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4731 35. Added mailbox_size and mailbox_filecount to appendfile.
4733 36. Added control = no_multiline_responses to ACLs.
4735 37. There was a bug in the logic of the code that waits for the clock to tick
4736 in the case where the clock went backwards by a substantial amount such
4737 that the microsecond fraction of "now" was more than the microsecond
4738 fraction of "then" (but the whole seconds number was less).
4740 38. Added support for the libradius Radius client library this is found on
4741 FreeBSD (previously only the radiusclient library was supported).
4747 1. When certain lookups returned multiple values in the form name=value, the
4748 quoting of the values was not always being done properly. Specifically:
4749 (a) If the value started with a double quote, but contained no whitespace,
4751 (b) If the value contained whitespace other than a space character (i.e.
4752 tabs or newlines or carriage returns) it was not quoted.
4753 This fix has been applied to the mysql and pgsql lookups by writing a
4754 separate quoting function and calling it from the lookup code. The fix
4755 should probably also be applied to nisplus, ibase and oracle lookups, but
4756 since I cannot test any of those, I have not disturbed their existing code.
4758 2. A hit in the callout cache for a specific address caused a log line with no
4759 reason for rejecting RCPT. Now it says "Previous (cached) callout
4760 verification failure".
4762 3. There was an off-by-one bug in the queryprogram router. An over-long
4763 return line was truncated at 256 instead of 255 characters, thereby
4764 overflowing its buffer with the terminating zero. As well as fixing this, I
4765 have increased the buffer size to 1024 (and made a note to document this).
4767 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4768 when Exim is waiting for an SMTP response from a remote server, Exim
4769 restarts its select() call on the socket, thereby resetting its timeout.
4770 This is not a problem when such interrupts are rare. Somebody set up a cron
4771 job to run exiwhat every 2 minutes, which is less than the normal select()
4772 timeout (5 or 10 minutes). This meant that the select() timeout never
4773 kicked in because it was always reset. I have fixed this by comparing the
4774 time when an interrupt arrives with the time at the start of the first call
4775 to select(). If more time than the timeout has elapsed, the interrupt is
4776 treated as a timeout.
4778 5. Some internal re-factoring in preparation for the addition of Sieve
4779 extensions (by MH). In particular, the "personal" test is moved to a
4780 separate function, and given an option for scanning Cc: and Bcc: (which is
4781 not set for Exim filters).
4783 6. When Exim created an email address using the login of the caller as the
4784 local part (e.g. when creating a From: or Sender: header line), it was not
4785 quoting the local part when it contained special characters such as @.
4787 7. Installed new OpenBSD configuration files.
4789 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4790 try to make them clearer.
4792 9. Callout options, other than the timeout value, were being ignored when
4793 verifying sender addresses in header lines. For example, when using
4795 verify = header_sender/callout=no_cache
4797 the cache was (incorrectly) being used.
4799 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4800 only the Cygwin environment.
4802 11. When return_path_on_delivery was set as a log selector, if different remote
4803 addresses in the same message used different return paths and parallel
4804 remote delivery occurred, the wrong values would sometimes be logged.
4805 (Whenever a remote delivery process finished, the return path value from
4806 the most recently started remote delivery process was logged.)
4808 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4809 lines when AUTH and/or TLS are in use. This is the "received protocol"
4810 field. Exim used to use "asmtp" for authenticated SMTP, without any
4811 indication (in the protocol name) for TLS use. Now it follows the RFC and
4812 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4813 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4814 names appear in log lines as well as in Received: header lines.
4816 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4817 extensions, and comparison tests, and to fix some bugs.
4819 14. Changes to the "personal" filter test:
4821 (1) The test was buggy in that it was just doing the equivalent of
4822 "contains" tests on header lines. For example, if a user's address was
4823 anne@some.where, the "personal" test would incorrectly be true for
4825 To: susanne@some.where
4827 This test is now done by extracting each address from the header in turn,
4828 and checking the entire address. Other tests that are part of "personal"
4829 are now done using regular expressions (for example, to check local parts
4830 of addresses in From: header lines).
4832 (2) The list of non-personal local parts in From: addresses has been
4833 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4834 taken from the Sieve specification recommendations.
4836 (3) If the message contains any header line starting with "List-" it is
4837 treated as non-personal.
4839 (4) The test for "circular" in the Subject: header line has been removed
4840 because it now seems ill-conceived.
4842 15. Minor typos in src/EDITME comments corrected.
4844 16. Installed latest exipick from John Jetmore.
4846 17. If headers_add on a router specified a text string that was too long for
4847 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4848 of string_sprintf() is now avoided.
4850 18. $message_body_size was not set (it was always zero) when running the DATA
4851 ACL and the local_scan() function.
4853 19. For the "mail" command in an Exim filter, no default was being set for
4854 the once_repeat time, causing a random time value to be used if "once" was
4855 specified. (If the value happened to be <= 0, no repeat happened.) The
4856 default is now 0s, meaning "never repeat". The "vacation" command was OK
4857 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4858 (I found it when inspecting the code).
4860 20. There is now an overall timeout for performing a callout verification. It
4861 defaults to 4 times the callout timeout, which applies to individual SMTP
4862 commands during the callout. The overall timeout applies when there is more
4863 than one host that can be tried. The timeout is checked before trying the
4864 next host. This prevents very long delays if there are a large number of
4865 hosts and all are timing out (e.g. when the network connections are timing
4866 out). The value of the overall timeout can be changed by specifying an
4867 additional sub-option for "callout", called "maxwait". For example:
4869 verify = sender/callout=5s,maxwait=20s
4871 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4872 to the end before writing, but this should make it even safer).
4874 22. Exim was forgetting that it had advertised PIPELINING for the second and
4875 subsequent messages on an SMTP connection. It was also not resetting its
4876 memory on STARTTLS and an internal HELO.
4878 23. When Exim logs an SMTP synchronization error within a session, it now
4879 records whether PIPELINING has been advertised or not.
4881 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4882 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4883 rather than long int.
4885 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4887 26. Added the never_mail option to autoreply.
4893 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4894 crash if the getsockname() call failed; this can happen if a connection is
4895 closed very soon after it is established. The problem was simply in the
4896 order in which certain operations were done, causing Exim to try to write
4897 to the SMTP stream before it had set up the file descriptor. The bug has
4898 been fixed by making things happen in the correct order.
4904 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4905 before the connection was closed, thus losing the rejection response.
4907 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4908 some early Solaris releases, but causes trouble in current releases where
4909 socklen_t is defined.
4911 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4914 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4917 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4918 that releasing the top of it at the end releases what was used for sub-
4919 expansions (unless the block got too big). However, discard this block if
4920 the first thing is a variable or header, so that we can use its block when
4921 it is dynamic (useful for very large $message_headers, for example).
4923 6. Lookups now cache *every* query, not just the most recent. A new, separate
4924 store pool is used for this. It can be recovered when all lookup caches are
4925 flushed. Lookups now release memory at the end of their result strings.
4926 This has involved some general refactoring of the lookup sources.
4928 7. Some code has been added to the store_xxx() functions to reduce the amount
4929 of flapping under certain conditions.
4931 8. log_incoming_interface used to affect only the <= reception log lines. Now
4932 it causes the local interface and port to be added to several more SMTP log
4933 lines, for example "SMTP connection from", and rejection lines.
4935 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4937 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4939 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4940 could be overwritten at the end of the current message (or the start of a
4941 new message if it was set in a HELO ACL). The value is now preserved for
4942 the duration of the SMTP connection.
4944 12. If a transport had a headers_rewrite setting, and a matching header line
4945 contained an unqualified address, that address was qualified, even if it
4946 did not match any rewriting rules. The underlying bug was that the values
4947 of the flags that permit the existence of unqualified sender and recipient
4948 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4949 non-local messages, and by -bnq for local messages) were not being
4950 preserved with the message after it was received.
4952 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4953 "next input=" as part of the text comprising the host identity instead of
4954 the correct text. The code was using the same buffer for two different
4955 strings. However, depending on which order the printing function evaluated
4956 its arguments, the bug did not always show up. Under Linux, for example, my
4957 test suite worked just fine.
4959 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4960 doesn't work with some older versions of Perl. It has been changed to "my",
4961 which in any case is probably the better facility to use.
4963 15. A really picky compiler found some instances of statements for creating
4964 error messages that either had too many or two few arguments for the format
4967 16. The size of the buffer for calls to the DNS resolver has been increased
4968 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4969 for addresses that have a lot of PTR records. This alleviates a problem; it
4970 does not fully solve it.
4972 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4973 buffer now truncates the list and logs the incident, which is the same
4974 action as happens when Exim is looking up a host name and its aliases.
4975 Previously in this situation something unpredictable would happen;
4976 sometimes it was "internal error: store_reset failed".
4978 18. If a server dropped the connection unexpectedly when an Exim client was
4979 using GnuTLS and trying to read a response, the client delivery process
4980 crashed while trying to generate an error log message.
4982 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4983 string, for example:
4985 warn message = H1: something\nH2: something
4987 the text was added as a single header line from Exim's point of view
4988 though it ended up OK in the delivered message. However, searching for the
4989 second and subsequent header lines using $h_h2: did not work. This has been
4990 fixed. Similarly, if a system filter added multiple headers in this way,
4991 the routers could not see them.
4993 20. Expanded the error message when iplsearch is called with an invalid key to
4994 suggest using net-iplsearch in a host list.
4996 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4997 ACLs are no longer actually imposed (and a message to that effect is
5000 22. If a "gecos" field in a passwd entry contained escaped characters, in
5001 particular, if it contained a \" sequence, Exim got it wrong when building
5002 a From: or a Sender: header from that name. A second bug also caused
5003 incorrect handling when an unquoted " was present following a character
5004 that needed quoting.
5006 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5007 was not being matched caselessly.
5009 24. Arranged for all hyphens in the exim.8 source to be escaped with
5012 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5013 properly. Recipient callout cache records were still being keyed to include
5014 the sender, even when use_sender was set false. This led to far more
5015 callouts that were necessary. The sender is no longer included in the key
5016 when use_sender is false.
5018 26. Added "control = submission" modifier to ACLs.
5020 27. Added the ${base62d: operator to decode base 62 numbers.
5022 28. dnsdb lookups can now access SRV records.
5024 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5025 the configuration file.
5027 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5028 (-v) mode. This makes the output for a verbose queue run more intelligible.
5030 31. Added a use_postmaster feature to recipient callouts.
5032 32. Added the $body_zerocount variable, containing the number of binary zero
5033 bytes in the message body.
5035 33. The time of last modification of the "new" subdirectory is now used as the
5036 "mailbox time last read" when there is a quota error for a maildir
5039 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5041 35. Added +ignore_unknown as a special item in host lists.
5043 36. Code for decoding IPv6 addresses in host lists is now included, even if
5044 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5045 address was recognized as an IP address, but was then not correctly decoded
5046 into binary, causing unexpected and incorrect effects when compared with
5053 1. Very minor rewording of debugging text in manualroute to say "list of
5054 hosts" instead of "hostlist".
5056 2. If verify=header_syntax was set, and a header line with an unqualified
5057 address (no domain) and a large number of spaces between the end of the
5058 name and the colon was received, the reception process suffered a buffer
5059 overflow, and (when I tested it) crashed. This was caused by some obsolete
5060 code that should have been removed. The fix is to remove it!
5062 3. When running in the test harness, delay a bit after writing a bounce
5063 message to get a bit more predictability in the log output.
5065 4. Added a call to search_tidyup() just before forking a reception process. In
5066 theory, someone could use a lookup in the expansion of smtp_accept_max_
5067 per_host which, without the tidyup, could leave open a database connection.
5069 5. Added the variables $recipient_data and $sender_data which get set from a
5070 lookup success in an ACL "recipients" or "senders" condition, or a router
5071 "senders" option, similar to $domain_data and $local_part_data.
5073 6. Moved the writing of debug_print from before to after the "senders" test
5076 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5077 problems for message scanning, either using a data ACL, or using
5078 local_scan() because the Received: header was not generated till after they
5079 were called (in order to set the time as the time of reception completion).
5080 I have revised the way this works. The header is now generated after the
5081 body is received, but before the ACL or local_scan() are called. After they
5082 are run, the timestamp in the header is updated.
5088 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5089 before starting a queue runner without re-exec. This happened only when
5090 deliver_drop_privilege was set or when the Exim user was set to root. The
5091 effect of the bug was that timeouts during subsequent deliveries caused
5092 crashes instead of being properly handled. The handler is now left at its
5093 default (and expected) setting.
5095 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5096 message, again when deliver_drop_privilege is set or Exim is run as root.
5097 The bug described in (1) was not present in this case, but the tidying up
5098 of the other signals was missing. I have made the two cases consistent.
5100 3. The ignore_target_hosts setting on a manualroute router was being ignored
5101 for hosts that were looked up using the /MX notation.
5103 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5106 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5107 operated on the sender address. After changing the $sender_address to <>
5108 for the sender address verify, Exim was re-instated it as the original
5109 (before rewriting) address, but remembering that it had rewritten it, so it
5110 wasn't rewriting it again. This bug also had the effect of breaking the
5111 sender address verification caching when the sender address was rewritten.
5113 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5114 This has been changed so that if the ip literal address matches
5115 ignore_target_hosts, the router declines.
5117 7. Added expansion conditions match_domain, match_address, and match_local_
5118 part (NOT match_host).
5120 8. The placeholder for the Received: header didn't have a length field set.
5122 9. Added code to Exim itself and to exim_lock to test for a specific race
5123 condition that could lead to file corruption when using MBX delivery. The
5124 issue is with the lockfile that is created in /tmp. If this file is removed
5125 after a process has opened it but before that process has acquired a lock,
5126 there is the potential for a second process to recreate the file and also
5127 acquire a lock. This could lead to two Exim processes writing to the file
5128 at the same time. The added code performs the same test as UW imapd; it
5129 checks after acquiring the lock that its file descriptor still refers to
5130 the same named file.
5132 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5133 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5136 11. Added the smtp_active_hostname option. If used, this will typically be made
5137 to depend on the incoming interface address. Because $interface_address is
5138 not set up until the daemon has forked a reception process, error responses
5139 that can happen earlier (such as "too many connections") no longer contain
5142 12. If an expansion in a condition on a "warn" statement fails because a lookup
5143 defers, the "warn" statement is abandoned, and the next ACL statement is
5144 processed. Previously this caused the whole ACL to be aborted.
5146 13. Added the iplsearch lookup type.
5148 14. Added ident_timeout as a log selector.
5150 15. Added tls_certificate_verified as a log selector.
5152 16. Added a global option tls_require_ciphers (compare the smtp transport
5153 option of the same name). This controls incoming TLS connections.
5155 17. I finally figured out how to make tls_require_ciphers do a similar thing
5156 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5157 before starting the TLS session.
5159 18. Tabs are now shown as \t in -bP output.
5161 19. If the log selector return_path_on_delivery was set, Exim crashed when
5162 bouncing a message because it had too many Received: header lines.
5164 20. If two routers both had headers_remove settings, and the first one included
5165 a superfluous trailing colon, the final name in the first list and the
5166 first name in the second list were incorrectly joined into one item (with a
5167 colon in the middle).
5173 1. Added -C and -D options to the exinext utility, mainly to make it easier
5174 to include in the automated testing, but these could be helpful when
5175 multiple configurations are in use.
5177 2. The exinext utility was not formatting the output nicely when there was
5178 an alternate port involved in the retry record key, nor when there was a
5179 message id as well (for retries that were specific to a specific message
5180 and a specific host). It was also confused by IPv6 addresses, because of
5181 the additional colons they contain. I have fixed the IPv4 problem, and
5182 patched it up to do a reasonable job for IPv6.
5184 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5185 delivery, the log line now contains "pipelined" if PIPELINING was used.
5187 4. An SMTP transport process used to panic and die if the bind() call to set
5188 an explicit outgoing interface failed. This has been changed; it is now
5189 treated in the same way as a connect() failure.
5191 5. A reference to $sender_host_name in the part of a conditional expansion
5192 that was being skipped was still causing a DNS lookup. This no longer
5195 6. The def: expansion condition was not recognizing references to header lines
5196 that used bh_ and bheader_.
5198 7. Added the _cache feature to named lists.
5200 8. The code for checking quota_filecount in the appendfile transport was
5201 allowing one more file than it should have been.
5203 9. For compatibility with Sendmail, the command line option
5211 and sets the incoming protocol and host name (for trusted callers). The
5212 host name and its colon can be omitted when only the protocol is to be set.
5213 Note the Exim already has two private options, -pd and -ps, that refer to
5214 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5215 "s", but I don't think that's a major issue.
5217 10. A number of refactoring changes to the code, none of which should affect
5220 (a) The number of logging options was getting close to filling up the
5221 32-bit word that was used as a bit map. I have split them into two classes:
5222 those that are passed in the argument to log_write(), and those that are
5223 only ever tested independently outside of that function. These are now in
5224 separate 32-bit words, so there is plenty of room for expansion again.
5225 There is no change in the user interface or the logging behaviour.
5227 (b) When building, for example, log lines, the code previously used a
5228 macro that called string_cat() twice, in order to add two strings. This is
5229 not really sufficiently general. Furthermore, there was one instance where
5230 it was actually wrong because one of the argument was used twice, and in
5231 one call a function was used. (As it happened, calling the function twice
5232 did not affect the overall behaviour.) The macro has been replaced by a
5233 function that can join an arbitrary number of extra strings onto a growing
5236 (c) The code for expansion conditions now uses a table and a binary chop
5237 instead of a serial search (which was left over from when there were very
5238 few conditions). Also, it now recognizes conditions like "pam" even when
5239 the relevant support is not compiled in: a suitably worded error message is
5240 given if an attempt is made to use such a condition.
5242 11. Added ${time_interval:xxxxx}.
5244 12. A bug was causing one of the ddress fields not to be passed back correctly
5245 from remote delivery subprocesses. The field in question was not being
5246 subsequently used, so this caused to problems in practice.
5248 13. Added new log selectors queue_time and deliver_time.
5250 14. Might have fixed a bug in maildirsizefile handling that threw up
5251 "unexpected character" debug warnings, and recalculated the data
5252 unnecessarily. In any case, I expanded the warning message to give more
5255 15. Added the message "Restricted characters in address" to the statements in
5256 the default ACL that block characters like @ and % in local parts.
5258 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5259 Three changes have been made:
5261 (a) There was a serious bug; a negative response to MAIL caused the whole
5262 recipient domain to be cached as invalid, thereby blocking all messages
5263 to all local parts at the same domain, from all senders. This bug has
5264 been fixed. The domain is no longer cached after a negative response to
5265 MAIL if the sender used is not empty.
5267 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5270 (c) A new callout option, "use_sender" has been added for people who want
5271 the modified behaviour.
5277 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5280 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5281 indeed breaks things for older releases.
5283 3. Added additional logging to the case where there is a problem reading data
5284 from a filter that is running in a subprocess using a pipe, in order to
5285 try to track down a specific problem.
5287 4. Testing facility fudge: when running in the test harness and attempting
5288 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5289 getting "No route to host". Convert this to a timeout.
5291 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5294 6. Some OS don't have socklen_t but use size_t instead. This affects the
5295 fifth argument of getsockopt() amongst other things. This is now
5296 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5297 can be set for individual OS. I have set it for SunOS5, OSF1, and
5298 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5299 some earlier ones do not.
5301 7. Change 4.30/15 was not doing the test caselessly.
5303 8. The standard form for an IPv6 address literal was being rejected by address
5304 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5305 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5306 this, as well as the form without the "IPv6" on the front (but only when
5307 address literals are enabled, of course).
5309 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5311 10. Exim crashed if a message with an empty sender address specified by -f
5312 encountered a router with an errors_to setting. This could be provoked only
5313 by a command such as
5317 where an empty string was supplied; "<>" did not hit this bug.
5319 11. Installed PCRE release 4.5.
5321 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5322 remained set. It is now erased.
5324 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5325 times from message ids (which are base 36 rather than the normal 62).
5327 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5328 were being counted as actual protocol errors, and logged if the log
5329 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5330 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5331 and DATA following a set of rejected RCPTs do not count as protocol errors.
5332 In other words, Exim assumes they were pipelined, though this may not
5333 actually be the case. Of course, in all cases the client gets an
5334 appropriate error code.
5336 15. If a lookup fails in an ACL condition, a message about the failure may
5337 be available; it is used if testing the ACL cannot continue, because most
5338 such messages specify what the cause of the deferral is. However, some
5339 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5340 that caused an old message to be retained and used if a later statement
5341 caused a defer, replacing the real cause of the deferral.
5343 16. If an IP address had so many PTR records that the DNS lookup buffer
5344 was not large enough to hold them, Exim could crash while trying to process
5345 the truncated data. It now detects and logs this case.
5347 17. Further to 4.21/58, another change has been made: if (and only if) the
5348 first line of a message (the first header line) ends with CRLF, a bare LF
5349 in a subsequent header line has a space inserted after it, so as not to
5350 terminate the header.
5352 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5353 unnecessarily, used atoi() instead of strtol(), and didn't check the
5354 termination when getting file sizes from file names by regex.
5356 19. Completely re-implemented the support for maildirsize files, in the light
5357 of a number of problems with the previous contributed implementation
5358 (4.30/29). In particular:
5360 . If the quota is zero, the maildirsize file is maintained, but no quota is
5363 . If the maildir directory does not exist, it is created before any attempt
5364 to write a maildirsize file.
5366 . The quota value in the file is just a cache; if the quota is changed in
5367 the transport, the new value overrides.
5369 . A regular expression is available for excluding directories from the
5372 20. The autoreply transport checks the characters in options that define the
5373 message's headers; it allows continued headers, but it was checking with
5374 isspace() after an embedded newline instead of explicitly looking for a
5377 21. If all the "regular" hosts to which an address was routed had passed their
5378 expiry times, and had not reached their retry times, the address was
5379 bounced, even if fallback hosts were defined. Now Exim should go on to try
5382 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5383 equivalent code in the SMTP transport. Some hosts send humungous responses
5384 to HELO/EHLO, more than 1024 it seems.
5386 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5387 gives compiler warnings in some environments. I've now done it "properly",
5390 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5391 (because of problems with the built-in one) was declared to return uschar *
5392 instead of char *, causing compiler failure.
5394 25. Fixed a file descriptor leak when processing alias/forward files.
5396 26. Fixed a minor format string issue in dbfn.c.
5398 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5400 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5401 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5404 29. When a sender address is verified, it is cached, to save repeating the test
5405 when there is more than one recipient in a message. However, when the
5406 verification involves a callout, it is possible for different callout
5407 options to be set for different recipients. It is too complicated to keep
5408 track of this in the cache, so now Exim always runs a verification when a
5409 callout is required, relying on the callout cache for the optimization.
5410 The overhead is duplication of the address routing, but this should not be
5413 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5414 to be verified with callout=postmaster, and the main callout worked but the
5415 postmaster check failed, the verification correctly failed. However, if a
5416 subsequent RCPT command asked for sender verification *without* the
5417 postmaster check, incorrect caching caused this verification also to fail,
5420 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5421 it was not caching the DNS options (qualify_single, search_parents) that
5422 were used when the lookup failed. A subsequent lookup with different
5423 options therefore always gave the same answer, though there were cases
5424 where it should not have. (Example: a "domains = !$mx_any" option on a
5425 dnslookup router: the "domains" option is always processed without any
5426 widening, but the router might have qualify_single set.) Now Exim uses the
5427 cached value only when the same options are set.
5429 32. Added John Jetmore's "exipick" utility to the distribution.
5431 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5432 than a timeout (e.g. a certificate is required, and is not provided), an
5433 Exim server now closes the connection immediately. Previously it waited for
5434 the client to close - but if the client is SSL, it seems that they each
5435 wait for each other, leading to a delay before one of them times out.
5437 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5438 maintained 0.8.x compatibility because I don't think many are using it, and
5439 it is clearly obsolete.
5441 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5444 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5445 client certificate was expired. A simple patch fixes this, though I don't
5446 understand the full logic of why the verify callback is called multiple
5449 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5450 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5451 which causes problems with some clients (such as the Certicom SSL Plus
5452 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5453 disables the coutermeasure allowing Eudora to connect."
5455 38. Exim was not checking that a write() to a log file succeeded. This could
5456 lead to Bad Things if a log got too big, in particular if it hit a file
5457 size limit. Exim now panics and dies if it cannot write to a log file, just
5458 as it does if it cannot open a log file.
5460 39. Modified OS/Makefile-Linux so that it now contains
5462 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5464 The two -D definitions ensure that Exim is compiled with large file
5465 support, which makes it possible to handle log files that are bigger than
5468 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5469 instance) a domain was checked against a named list that involved a lookup,
5470 causing $domain_data to be set, then another domain was checked against the
5471 same list, then the first domain was re-checked, the value of $domain_data
5472 after the final check could be wrong. In particular, if the second check
5473 failed, it could be set empty. This bug probably also applied to
5476 41. The strip_trailing_dot option was not being applied to the address given
5477 with the -f command-line option.
5479 42. The code for reading a message's header from the spool was incrementing
5480 $received_count, but never initializing it. This meant that the value was
5481 incorrect (doubled) while delivering a message in the same process in which
5482 it was received. In the most common configuration of Exim, this never
5483 happens - a fresh exec is done - but it can happen when
5484 deliver_drop_privilege is set.
5486 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5487 it now includes up to 150 characters of the unexpected data in the log
5490 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5491 and building data strings. The size of both of these buffers was 10 000
5492 bytes - far larger than anybody would *ever* want, thought I. Needless to
5493 say, somebody hit the limit. I have increased the maximum line length to
5494 20 000 and the maximum data length of concatenated lines to 100 000. I have
5495 also fixed two bugs, because there was no checking on these buffers. Tsk,
5496 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5497 buffer is too small.
5499 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5500 lsearch lookups. Now it does.
5502 46. When parsing a route_list item in a manualroute router, a fixed-length
5503 buffer was used for the list of hosts. I made this 1024 bytes long,
5504 thinking that nobody would ever have a list of hosts that long. Wrong.
5505 Somebody had a whole pile of complicated expansion conditions, and the
5506 string was silently truncated, leading to an expansion error. It turns out
5507 that it is easier to change to an unlimited length (owing to other changes
5508 that have happened since this code was originally written) than to build
5509 structure for giving a limitation error. The length of the item that
5510 expands into the list of hosts is now unlimited.
5512 47. The lsearch lookup could not handle data where the length of text line was
5513 more than 4095 characters. Such lines were truncated, leading to shortened
5514 data being returned. It should now handle lines of any length.
5516 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5517 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5520 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5521 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5522 no longer happens in comment lines. A list of replacements is now placed
5523 at the head of all of the source files, except those whose only change is
5524 to replace PERL_COMMAND in the very first #! line.
5526 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5527 messages on the queue, with a bottom-up merge sort, using code contributed
5528 by Michael Haardt. This should make operations like -bp somewhat faster on
5529 large queues. It won't affect queue runners, except when queue_run_in_order
5532 51. Installed eximstats 1.31 in the distribution.
5534 52. Added support for SRV lookups to the dnslookup router.
5536 53. If an ACL referred to $message_body or $message_body_end, the value was not
5537 reset for any messages that followed in the same SMTP session.
5539 54. The store-handling optimization for building very long strings was not
5540 differentiating between the different store pools. I don't think this
5541 actually made any difference in practice, but I've tidied it.
5543 55. While running the routers to verify a sender address, $sender_address
5544 was still set to the sender address. This is wrong, because when routing to
5545 send a bounce to the sender, it would be empty. Therefore, I have changed
5546 it so that, while verifying a sender address, $sender_address is set to <>.
5547 (There is no change to what happens when verifying a recipient address.)
5549 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5550 target A or AAAA records (if not already returned) without resetting the
5551 qualify_single or search_parents options of the DNS resolver. These are
5552 inappropriate in this case because the targets of MX and SRV records must
5553 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5554 target that, when qualified, matched something in the local domain. These
5555 two options are now turned off when doing these lookups.
5557 57. It seems that at least some releases of Reiserfs (which does not have the
5558 concept of a fixed number of inodes) returns zero and not -1 for the
5559 number of available inodes. This interacted badly with check_spool_inodes,
5560 which assumed that -1 was the "no such thing" setting. What I have done is
5561 to check that the total number of inodes is greater than zero before doing
5562 the test of how many are available.
5564 58. When a "warn" ACL statement has a log_message modifier, the message is
5565 remembered, and not repeated. This is to avoid a lot of repetition when a
5566 message has many recipients that cause the same warning to be written.
5567 Howewer, Exim was preserving the list of already written lines for an
5568 entire SMTP session, which doesn't seem right. The memory is now reset if a
5569 new message is started.
5571 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5572 debugging output unless log_rewrite was also set.
5574 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5575 of (int)(handle) when we know that handle contains (void *)(-1).
5577 61. The Exim daemon panic-logs an error return when it closes the incoming
5578 connection. However "connection reset by peer" seems to be common, and
5579 isn't really an error worthy of noting specially, so that particular error
5582 62. When Exim is trying to find all the local interfaces, it used to panic and
5583 die if the ioctl to get the interface flags failed. However, it seems that
5584 on at least one OS (Solaris 9) it is possible to have an interface that is
5585 included in the list of interfaces, but for which you get a failure error
5586 for this call. This happens when the interface is not "plumbed" into a
5587 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5588 failure of the "get flags" call assumes that the interface is down.
5590 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5591 makes life easier for people who are doing arithmetic on fields extracted
5592 from dates, where you often get leading zeros that should not be
5593 interpreted as octal.
5595 64. Added qualify_domain to the redirect router, to override the global
5598 65. If a pathologically long header line contained very many addresses (the
5599 report of this problem mentioned 10 000) and each of them was rewritten,
5600 Exim could use up a very large amount of memory. (It kept on making new
5601 copies of the header line as it rewrote, and never released the old ones.)
5602 At the expense of a bit more processing, the header rewriting function has
5603 been changed so that it no longer eats memory in this way.
5605 66. The generation of the Received: header has been moved from the time that a
5606 message starts to be received, to the time that it finishes. The timestamp
5607 in the Received: header should now be very close to that of the <= log
5608 line. There are two side-effects of this change:
5610 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5611 logged header lines no longer include the local Received: line, because
5612 it has not yet been created. The same applies to a copy of the message
5613 that is returned to a non-SMTP sender when a message is rejected.
5615 (b) When a filter file is tested using -bf, no additional Received: header
5616 is added to the test message. After some thought, I decided that this
5619 This change does not affect the value of $received_for. It is still set
5620 after address rewriting, but before local_scan() is called.
5622 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5624 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5625 gave an unhelpful panic error message, and a defer error. I have managed to
5626 change this behaviour so that it now rejects any supplied certificate,
5627 which seems right, as the list of acceptable certificates is empty.
5629 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5630 gave an unhelpful defer error. I have not managed to make this reject any
5631 supplied certificates, but the error message it gives is "no certificate
5632 supplied", which is not helpful.
5634 70. exigrep's output now also includes lines that are not associated with any
5635 message, but which match the given pattern. Implemented by a patch from
5636 Martin Sluka, which also tidied up the Perl a bit.
5638 71. Recipient callout verification, like sender verification, was using <> in
5639 the MAIL FROM command. This isn't really the right thing, since the actual
5640 sender may affect whether the remote host accepts the recipient or not. I
5641 have changed it to use the actual sender in the callout; this means that
5642 the cache record is now keyed on a recipient/sender pair, not just the
5643 recipient address. There doesn't seem to be a real danger of callout loops,
5644 since a callout by the remote host to check the sender would use <>.
5645 [SEE ABOVE: changed after hitting problems.]
5647 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5648 temporary errors. However, in the case of such a code being given after
5649 the end of a data transmission (i.e. after ".") Exim was failing to write
5650 a retry record for the message. (Yes, there was some broken host that was
5651 actually sending 8xx at this point.)
5653 73. An unknown lookup type in a host list could cause Exim to panic-die when
5654 the list was checked. (An example that provoked this was putting <; in the
5655 middle of a list instead of at the start.) If this happened during a DATA
5656 ACL check, a -D file could be left lying around. This kind of configuration
5657 error no longer causes Exim to die; instead it causes a defer errror. The
5658 incident is still logged to the main and panic logs.
5660 74. Buglet left over from Exim 3 conversion. The message "too many messages
5661 in one connection" was written to the rejectlog but not the mainlog, except
5662 when address rewriting (yes!) was being logged.
5664 75. Added write_rejectlog option.
5666 76. When a system filter was run not as root (that is, when system_filter_user
5667 was set), the values of the $n variables were not being returned to the
5668 main process; thus, they were not subsequently available in the $sn
5671 77. Added +return_path_on_delivery log selector.
5673 78. A connection timeout was being treated differently from recipients deferred
5674 when testing hosts_max_try with a message that was older than the host's
5675 retry timeout. (The host should not be counted, thus allowing all hosts to
5676 be tried at least once before bouncing.) This may have been the cause of an
5677 occasionally reported bug whereby a message would remain on the queue
5678 longer than the retry timeout, but would be bounced if a delivery was
5679 forced. I say "may" because I never totally pinned down the problem;
5680 setting up timeout/retry tests is difficult. See also the next item.
5682 79. The ultimate address timeout was not being applied to errors that involved
5683 a combination of host plus message (for example, a timeout on a MAIL
5684 command). When an address resolved to a number of possible hosts, and they
5685 were not all tried for each delivery (e.g. because of hosts_max_try), a
5686 message could remain on the queue longer than the retry timeout.
5688 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5691 81. Fixed an obscure SMTP outgoing bug which required at least the following
5692 conditions: (a) there was another message waiting for the same server;
5693 (b) the server returned 5xx to all RCPT commands in the first message so
5694 that the message was not completed; (c) the server dropped the connection
5695 or gave a negative response to the RSET that Exim sends to abort the
5696 transaction. The observed case was a dropped connection after DATA that had
5697 been sent in pipelining mode. That is, the server had advertised PIPELINING
5698 but was not implementing it correctly. The effect of the bug was incorrect
5699 behaviour, such as trying another host, and this could lead to a crash.
5705 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5706 and daemon.c were passed as pointers to ints; they should have been
5707 pointers to socklen_t variables (which are typically unsigned ints).
5709 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5712 3. Fixed a really odd bug that affected only the testing scheme; patching a
5713 certain fixed string in the binary changed the value of another string that
5714 happened to be identical to the end of the original first string.
5716 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5717 name", it returns that address as the IP address. On some operating
5718 systems (e.g. Solaris), it also passes back the IP address string as the
5719 "host name". However, on others (e.g. Linux), it passes back an empty
5720 string. Exim wasn't checking for this, and was changing the host name to an
5721 empty string, assuming it had been canonicized.
5723 5. Although rare, it is permitted to have more than one PTR record for a given
5724 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5725 all the names associated with an address, because they do in Solaris.
5726 However, it seems that they do not in Linux for data that comes from the
5727 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5728 I found this out when I moved to a new Linux workstation and tried to run
5729 the Exim test suite.
5731 To get round this problem I have changed the code so that it now does its
5732 own call to the DNS to look up PTR records when searching for a host name.
5733 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5734 addresses that are only in /etc/hosts are still found.
5736 This behaviour is, however, controlled by an option called host_lookup_
5737 order, which defaults to "bydns:byaddr". If people want to use the other
5738 order, or indeed, just use one or the other means of lookup, they can
5739 specify it in this variable.
5741 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5742 some operating systems, this comes back from gethostbyaddr() as an empty
5743 string, and this is what Exim used to test for. However, it seems that in
5744 other systems, "." is yielded. Exim now tests for this case too.
5746 7. The values of check_spool_space and check_log_space are now held internally
5747 as a number of kilobytes instead of an absolute number of bytes. If a
5748 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5749 kilobyte. This means that much larger values can be stored.
5751 8. Exim monitor: an attempt to get the action menu when not actually pointing
5752 at a message produces an empty menu entitled "No message selected". This
5753 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5754 no entries in it ("Shell widget menu has zero width and/or height"). So I
5755 have added a single, blank menu entry in this case.
5757 9. Added ${quote_local_part.
5759 10. MIME decoding is now applied to the contents of Subject: header lines when
5762 11. Now that a reference to $sender_host_address automatically causes a reverse
5763 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5764 host lookup before query-style lookups in lists that might use this
5765 variable. This has therefore been abolished, and the "net-" prefix is no
5766 longer necessary for query-style lookups.
5768 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5769 have been a typo for LFLAGS, so it has been changed.
5771 13. The install script calls Exim with "-C /dev/null" in order to find the
5772 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5773 to be output. Howeve, since Exim outputs its version number before the
5774 error, it didn't break the script. It just looked ugly. I fixed this by
5775 always allowing "-C /dev/null" if the caller is root.
5777 14. Ignore overlarge ACL variable number when reading spool file - insurance
5778 against a later release with more variables having written the file.
5780 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5781 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5782 this, as well as the form without the "IPv6" on the front.
5784 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5785 OS/Makefile-Darwin file.
5787 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5788 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5790 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5791 variable. However, some OS use ints for inodes. I've added cast to long int
5792 to get rid of the compiler warning.
5794 19. I had forgotten to lock out "/../" in configuration file names when
5795 ALT_CONFIG_PREFIX was set.
5797 20. Routers used for verification do not need to specify transports. However,
5798 if such a router generated a host list, and callout was configured, Exim
5799 crashed, because it could not find a port number from the (non-existent)
5800 transport. It now assumes port 25 in this circumstance.
5802 21. Added the -t option to exigrep.
5804 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5805 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5806 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5807 them is set, LOOKUP_LSEARCH is forced.
5809 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5811 24. Added sender and host information to the "rejected by local_scan()" log
5812 line; previously there was no indication of these.
5814 25. Added .include_if_exists.
5816 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5817 Linux. It turns out that not all file systems support this. Apparently some
5818 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5819 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5820 not supported on the file descriptor, is now ignored when Exim is trying to
5821 sync a directory. This applies only to Linux.
5823 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5825 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5826 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5829 29. Added support for maildirsize files from supplied patch (modified a bit).
5831 30. The use of :fail: followed by an empty string could lead Exim to respond to
5832 sender verification failures with (e.g.):
5834 550 Verification failed for <xxx>
5835 550 Sender verify failed
5837 where the first response line was missing the '-' that indicates it is not
5838 the final line of the response.
5840 31. The loop for finding the name of the user that called Exim had a hardwired
5841 limit of 10; it now uses the value of finduser_retries, which is used for
5842 all other user lookups.
5844 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5847 33. Exim was neglecting to zero errno before one call of strtol() when
5848 expanding a string and expecting an integer value. On some systems this
5849 resulted in spurious "integer overflow" errors. Also, it was casting the
5850 result into an int without checking.
5852 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5853 did not work. The code looks as if it has *never* worked, though it appears
5854 to have been documented since at least releast 1.62. I have made it work.
5856 35. The "timeout_DNS" error in retry rules, also documented since at least
5857 1.62, also never worked. As it isn't clear exactly what this means, and
5858 clearly it isn't a major issue, I have abolished the feature by treating it
5859 as "timeout", and writing a warning to the main and panic logs.
5861 36. The display of retry rules for -brt wasn't always showing the error code
5864 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5865 timeout_connect_A, timeout_connect_MX.
5867 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5868 to the empty sender.
5870 39. The daemon was not analysing the content of -oX till after it had closed
5871 stderr and disconnected from the controlling terminal. This meant that any
5872 syntax errors were only noted on the panic log, and the return code from
5873 the command was 0. By re-arranging the code a little, I've made the
5874 decoding happen first, so such errors now appear on stderr, and the return
5875 code is 1. However, the actual setting up of the sockets still happens in
5876 the disconnected process, so errors there are still only recorded on the
5879 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5880 connections (as happens on some IP stacks) was logged at start up time as
5881 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5882 it from "IPv6 and IPv4", which means that two separate sockets are being
5885 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5886 says whether AF_INET or AF_INET6 was passed as an argument.
5888 42. Exiwhat output was messed up when time zones were included in log
5891 43. Exiwhat now gives more information about the daemon's listening ports,
5892 and whether -tls-on-connect was used.
5894 44. The "port" option of the smtp transport is now expanded.
5896 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5897 silently ignored. Now an error message is written to the main and panic
5900 46. There's a new ACL modifier called "logwrite" which writes to a log file
5901 as soon as it is encountered.
5903 47. Added $local_user_uid and $local_user_gid at routing time.
5905 48. Exim crashed when trying to verify a sender address that was being
5908 49. Exim was recognizing only a space character after ".include". It now also
5909 recognizes a tab character.
5911 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5912 extracting the relevant information from the specification. The man page no
5913 longer contains scrambled data for the -d option, and I've added a section
5914 at the front about calling Exim under different names.
5916 51. Added "extra_headers" argument to the "mail" command in filter files.
5918 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5921 53. Installed eximstats 1.29.
5923 54. Added transport_filter_timeout as a generic transport option.
5925 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5926 Cc: header lines. This was required by RFC 822, but it not required by RFC
5929 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5930 incoming messages that did not have them. Now it does so only if the
5931 message originates locally, that is, if there is no associated remote host
5932 address. When Resent- header lines are present, this applies to the Resent-
5933 lines rather than the non-Resent- lines.
5935 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5936 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5938 58. Messages for configuration errors now include the name of the main
5939 configuration file - useful now that there may be more than one file in a
5940 list (.included file names were always shown).
5942 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5943 for those rare installations that do not start the daemon as root or run it
5944 setuid root. I've cut out the call to initgroups() if the daemon is not
5947 60. The Exim user and group can now be bound into the binary as text strings
5948 that are looked up at the start of Exim's processing.
5950 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5952 62. Added $mailstore_basename variable.
5954 63. Installed patch to sieve.c from Michael Haardt.
5956 64. When Exim failed to open the panic log after failing to open the main log,
5957 the original message it was trying to log was written to stderr and debug
5958 output, but if they were not available (the usual case in production), it
5959 was lost. Now it is written to syslog before the two lines that record the
5960 failures to open the logs.
5962 65. Users' Exim filters run in subprocesses under the user's uid. It is
5963 possible for a "deliver" command or an alias in a "personal" command to
5964 provoke an address rewrite. If logging of address rewriting is configured,
5965 this fails because the process is not running as root or exim. There may be
5966 a better way of dealing with this, but for the moment (because 4.30 needs
5967 to be released), I have disabled address rewrite logging when running a
5968 filter in a non-root, non-exim process.
5974 1. The buildconfig auxiliary program wasn't quoting the value set for
5975 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5976 not defined. This bug was masked in 4.22 by the effect that was fixed in
5979 2. Some messages that were rejected after a message id was allocated were
5980 shown as "incomplete" by exigrep. It no longer does this for messages that
5981 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5983 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5984 have allow_domain_literals set, the ID did not get logged in the <= line.
5985 Domain literals are now always recognized in Message-ID: header lines.
5987 4. The first argument for a ${extract expansion item is the key name or field
5988 number. Leading and trailing spaces in this item were not being ignored,
5989 causing some misleading effects.
5991 5. When deliver_drop_privilege was set, single queue runner processes started
5992 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5993 same command in the process it spins off) were not dropping privilege.
5995 6. When the daemon running as "exim" started a queue runner, it always
5996 re-executed Exim in the spun-off process. This is a waste of effort when
5997 deliver_drop_privilege is set. The new process now just calls the
5998 queue-runner function directly.
6004 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6007 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6008 ignored. Though the use of -oP was forcing the writing of a pid file, it
6009 was always written to the default place.
6011 3. If the message "no IP address found for host xxxx" is generated during
6012 incoming verification, it is now followed by identification of the incoming
6013 connection (so you can more easily find what provoked it).
6015 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6017 5. Added some features to "harden" Exim a bit more against certain attacks:
6019 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6020 be put in Local/Makefile. This is like the never_users runtime option,
6021 but it cannot be overridden. The default setting is "root".
6023 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6024 prefix string with which any file named in a -C command line option
6027 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6028 is retained for -C and -D only if the caller of Exim is root. Without
6029 it, the exim user may also use -C and -D and retain privilege.
6031 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6032 command line option is disabled.
6034 6. Macro names set by the -D option must start with an upper case letter, just
6035 like macro names defined in the configuration file.
6037 7. Added "dereference=" facility to LDAP.
6039 8. Two instances of the typo "uknown" in the source files are fixed.
6041 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6042 the Configure-Makefile script screwed up while processing it.
6044 10. Incorporated PCRE 4.4.
6046 11. The SMTP synchronization check was not operating right at the start of an
6047 SMTP session. For example, it could not catch a HELO sent before the client
6048 waited for the greeting. There is now a check for outstanding input at the
6049 point when the greeting is written. Because of the duplex, asynchronous
6050 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6051 way, but not yet received, when the check is performed.
6053 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6054 on TCP/IP sockets, because this apparently causes some broken clients to
6057 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6058 unchanged) from the Cygwin maintainer.
6060 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6061 was supported instead of testing for mbx. Effectively a typo.
6063 15. The spa authenticator server code was not checking that the input it
6064 received was valid base64.
6066 16. The debug output line for the "set" modifier in ACLs was not showing the
6067 name of the variable that was being set.
6069 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6071 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6072 lookup on its own. Something else was needed to trigger the lookup. For
6073 example, a match in host_lookup or the need for a host name in a host list.
6074 Now, if $sender_host_name is referenced and the host name has not yet been
6075 looked up, a lookup is performed. If the lookup fails, the variable remains
6076 empty, and $host_lookup_failed is set to "1".
6078 19. Added "eqi" as a case-independent comparison operator.
6080 20. The saslauthd authentication condition could segfault if neither service
6081 nor realm was specified.
6083 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6084 error message that was logged was misleading, and incoming SMTP
6085 connections were dropped. The message is now more accurate, and temporary
6086 errors are given to SMTP connections.
6088 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6089 (except RSET) with a 421 error, until QUIT is received. However, it was
6090 failing to send a response to QUIT.
6092 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6093 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6094 ACL is now run after the helo_try_verify_hosts code.
6096 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6097 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6098 case-independent, so other case variants are also recognized). Apparently
6099 some systems use these upper case variants.
6101 25. If more than two messages were waiting for the same host, and a transport
6102 filter was specified for the transport, Exim sent two messages over the
6103 same TCP/IP connection, and then failed with "socket operation on non-
6104 socket" when it tried to send the third.
6106 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6108 27. The extern definition of crypt16() in expand.c was not being excluded when
6109 the OS had its own crypt16() function.
6111 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6112 as a preferred synonym for return_size_limit, both as an option and as an
6115 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6117 30. Changed the default configuration ACL to relax the local part checking rule
6118 for addresses that are not in any local domains. For these addresses,
6119 slashes and pipe symbols are allowed within local parts, but the sequence
6120 /../ is explicitly forbidden.
6122 31. SPA server authentication was not clearing the challenge buffer before
6125 32. log_message in a "warn" ACL statement was writing to the reject log as
6126 well as to the main log, which contradicts the documentation and doesn't
6127 seem right (because no rejection is happening). So I have stopped it.
6129 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6130 However, I am unable to do any testing of this.
6132 34. Fixed an infelicity in the appendfile transport. When checking directories
6133 for a mailbox, to see if any needed to be created, it was accidentally
6134 using path names with one or more superfluous leading slashes; tracing
6135 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6137 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6138 contents are added to the log line that is written for every discarded
6139 recipient. (Previously a log_message setting was ignored.)
6141 36. The ${quote: operator now quotes the string if it is empty.
6143 37. The install script runs exim in order to find its version number. If for
6144 some reason other than non-existence or emptiness, which it checks, it
6145 could not run './exim', it was installing it with an empty version number,
6146 i.e. as "exim-". This error state is now caught, and the installation is
6149 38. An argument was missing from the function that creates an error message
6150 when Exim fails to connect to the socket for saslauthd authentication.
6151 This could cause Exim to crash, or give a corrupted message.
6153 39. Added isip, isip4, and isip6 to ${if conditions.
6155 40. The ACL variables $acl_xx are now saved with the message, and can be
6156 accessed later in routers, transports, and filters.
6158 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6159 strings in the file are not string-expanded.
6161 42. If a MAIL command specified a SIZE value that was too large to fit into an
6162 int variable, the check against message_size_limit failed. Such values are
6163 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6164 day this will have to be increased, but I don't think I want to be around
6165 when emails are that large.
6172 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6173 iconv() is not standard in FreeBSD.
6175 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6176 IPv6 enabled. The observed symptom was a segmentation fault on return from
6177 the function os_common_find_running_interfaces() in src/os.c.
6179 3. In the check_special_case() function in daemon.c I had used "errno" as an
6180 argument name, which causes warnings on some systems. This was basically a
6181 typo, since it was named "eno" in the comments!
6183 4. The code that waits for the clock to tick (at a resolution of some fraction
6184 of a second) so as to ensure message-id uniqueness was always waiting for
6185 at least one whole tick, when it could have waited for less. [This is
6186 almost certainly not relevant at current processor speeds, where it is
6187 unlikely to ever wait at all. But we try to future-proof.]
6189 5. The function that sleeps for a time interval that includes fractions of a
6190 second contained a race. It did not block SIGALRM between setting the
6191 timer, and suspending (a couple of lines later). If the interval was short
6192 and the sigsuspend() was delayed until after it had expired, the suspension
6193 never ended. On busy systems this could lead to processes getting stuck for
6196 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6197 process, before it forks any delivery processes. The open lookup caching
6198 mechanism meant that the open file or database connection was passed into
6199 the delivery process. The problem was that delivery processes always tidy
6200 up cached lookup data. This could cause a problem for the next delivery
6201 process started by the queue runner, because the external queue runner
6202 process does not know about the closure. So the next delivery process
6203 still has data in the lookup cache. In the case of a file lookup, there was
6204 no problem because closing a file descriptor in a subprocess doesn't affect
6205 the parent. However, if the lookup was caching a connection to a database,
6206 the connection was closed, and the second delivery process was likely to
6207 see errors such as "PGSQL: query failed: server closed the connection
6208 unexpectedly". The problem has been fixed by closing all cached lookups
6209 in a queue runner before running a delivery process.
6211 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6212 seem to have the "const" qualifier which it has on other OS. I've
6215 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6216 *of the same type* (client or server) with the same public name that an
6217 error should be diagnosed.
6219 9. When Exim looked up a host name for an IP address, but failed to find the
6220 original IP address when looking up the host name (a safety check), it
6221 output the message "<ip address> does not match any IP for NULL", which was
6222 confusing, to say the least. The bug was that the host name should have
6223 appeared instead of "NULL".
6225 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6226 user that is built into the binary, and the -C or -D options is used, root
6227 privilege is dropped before the configuration file is read. In addition,
6228 logging is switched to stderr instead of the normal log files. If the
6229 configuration then re-defines the Exim user, the unprivileged environment
6230 is probably not what is expected, so Exim logs a panic warning message (but
6233 However, if deliver_drop_privilege is set, the unprivileged state may well
6234 be exactly what is intended, so the warning has been cut out in that case,
6235 and Exim is allowed to try to write to its normal log files.
6241 1. smtp_return_error_details was not giving details for temporary sender
6242 or receiver verification errors.
6244 2. Diagnose a configuration error if two authenticators have the same public
6247 3. Exim used not to create the message log file for a message until the first
6248 delivery attempt. This could be confusing when incoming messages were held
6249 for policy or load reasons. The message log file is now created at the time
6250 the message is received, and an initial "Received" line is written to it.
6252 4. The automatically generated man page for command line options had a minor
6253 bug that caused no ill effects; however, a more serious problem was that
6254 the procedure for building the man page automatically didn't always
6255 operate. Consequently, release 4.20 contains an out-of-date version. This
6256 shouldn't happen again.
6258 5. When building Exim with embedded Perl support, the script that builds the
6259 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6260 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6262 6. The freeze_tell option was not being used for messages that were frozen on
6263 arrival, either by an ACL or by local_scan().
6265 7. Added the smtp_incomplete_transaction log selector.
6267 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6268 was accepting AUTH without a new EHLO.
6270 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6271 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6274 10. Logging of TCP/IP connections (when configured) now happens in the main
6275 daemon process instead of the child process, so that the TCP/IP connection
6276 count is more accurate (but it can never be perfect).
6278 11. The use of "drop" in a nested ACL was not being handled correctly in the
6279 outer ACL. Now, if condition failure induced by the nested "drop" causes
6280 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6281 or "require"), the connection is dropped.
6283 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6284 that yield "discard" can now be used with an "accept" or a "discard" verb,
6285 but an error is generated for any others (because I can't see a useful way
6286 to define what should happen).
6288 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6289 are now processed in the same way as lines in the Exim configuration file.
6290 In particular, continuation lines are supported.
6292 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6294 15. Added -ti meaning -t -i.
6296 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6297 domains, and warn by logging if others are found.
6299 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6300 returned from GIFCONF when Exim is trying to find the list of interfaces on
6301 a host. The code in os.c has been modified to copy each ifreq to an aligned
6302 structure in all cases.
6304 Also, in some cases, the returned ifreq's were being copied to a 'struct
6305 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6306 means the last couple of bytes of an IPv6 address could be chopped if the
6307 ifreq contained only a normal sockaddr (14 bytes storage).
6309 18. Named domain lists were not supported in the hosts_treat_as_local option.
6310 An entry such as +xxxx was not recognized, and was treated as a literal
6313 19. Ensure that header lines added by a DATA ACL are included in the reject log
6314 if the ACL subsequently rejects the message.
6316 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6317 MD5 (which is deprecated).
6319 21. When testing a filter file using -bf, Exim was writing a message when it
6320 took the sender from a "From " line in the message, but it was not doing so
6321 when it took $return_path from a Return-Path: header line. It now does.
6323 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6324 with a valid header line field name (a series of printing characters
6325 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6327 23. Changed "disc" in the source to "disk" to conform to the documentation and
6328 the book and for uniformity.
6330 24. Ignore Sendmail's -Ooption=value command line item.
6332 25. When execve() failed while trying to run a command in a pipe transport,
6333 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6334 could be confused with a return value of 69 from the command itself. This
6335 has been changed to 127, the value the shell returns if it is asked to run
6336 a non-existent command. The wording for the related log line suggests a
6337 non-existent command as the problem.
6339 26. If received_header_text expands to an empty string, do not add a Received:
6340 header line to the message. (Well, it adds a token one on the spool, but
6341 marks it "old" so that it doesn't get used or transmitted.)
6343 27. Installed eximstats 1.28 (addition of -nt option).
6345 28. There was no check for failure on the call to getsockname() in the daemon
6346 code. This can fail if there is a shortage of resources on the system, with
6347 ENOMEM, for example. A temporary error is now given on failure.
6349 29. Contrary to the C standard, it seems that in some environments, the
6350 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6351 program. Exim now does this explicitly; it affects the formatting of
6352 timestamps using strftime().
6354 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6355 complaints. I've now initialized all the variables, to avoid this.
6357 32. Header lines added by a system filter were not being "seen" during
6358 transport-time rewrites.
6360 33. The info_callback() function passed to OpenSSL is set up with type void
6361 (*)(SSL *, int, int), as described somewhere. However, when calling the
6362 function (actually a macro) that sets it up, the type void(*)() is
6363 expected. I've put in a cast to prevent warnings from picky compilers.
6365 34. If a DNS black list lookup found a CNAME record, but there were no A
6366 records associated with the domain it pointed at, Exim crashed.
6368 35. If a DNS black list lookup returned more than one A record, Exim ignored
6369 all but the first. It now scans all returned addresses if a particular IP
6370 value is being sought. In this situation, the contents of the
6371 $dnslist_value variable are a list of all the addresses, separated by a
6374 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6375 to accept a host name and all its aliases if the forward lookup for any of
6376 them yielded the IP address of the incoming connection. Now it accepts only
6377 those names whose forward lookup yields the correct IP address. Any other
6378 names are discarded. This closes a loophole whereby a rogue DNS
6379 administrator could create reverse DNS records to break through a
6380 wildcarded host restriction in an ACL.
6382 37. If a user filter or a system filter that ran in a subprocess used any of
6383 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6384 the wrong values were passed to the pipe command ($thisaddress had the
6385 value of $0, $0 had the value of $1, etc). This bug was introduced by
6386 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6388 38. Improved the line breaking for long SMTP error messages from ACLs.
6389 Previously, if there was no break point between 40 and 75 characters, Exim
6390 left the rest of the message alone. Two changes have been made: (a) I've
6391 reduced the minimum length to 35 characters; (b) if it can't find a break
6392 point between 35 and 75 characters, it looks ahead and uses the first one
6393 that it finds. This may give the occasional overlong line, but at least the
6394 remaining text gets split now.
6396 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6397 file descriptors might be low, and that setting 1000 would always raise it.
6398 It turns out that in some environments, the limit is already over 1000 and
6399 that lowering it causes trouble. So now Exim takes care not to decrease it.
6401 40. When delivering a message, the value of $return_path is set to $sender_
6402 address at the start of routing (routers may change the value). By an
6403 oversight, this default was not being set up when an address was tested by
6404 -bt or -bv, which affected the outcome if any router or filter referred to
6407 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6408 log only when "message" or "log_message" are set. However, if one of the
6409 conditions was an address verification, or a call to a nested ACL, the
6410 messages generated by the underlying test were being passed through. This
6411 no longer happens. The underlying message is available in $acl_verify_
6412 message for both "message" and "log_message" expansions, so it can be
6413 passed through if needed.
6415 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6416 new expansion $bh_ to give the encoded byte string without charset
6417 translation. Translation happens only if iconv() is available; HAVE_ICONV
6418 indicates this at build time. HEADERS_CHARSET gives the charset to
6419 translate to; headers_charset can change it in the configuration, and
6420 "headers charset" can change it in an individual filter file.
6422 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6423 that creates RFC 2047 encoded "words" labels them as that charset instead
6424 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6425 expansion operator; (ii) when Exim creates a From: line for a local
6426 message; (iii) when a header line is rewritten to include a "phrase" part.
6428 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6429 buggy, causing it to skip the first lines of messages whose message ID
6430 ended in 'D'. This would not have bitten before Exim release 4.14, because
6431 message IDs were unlikely to end in 'D' before then. The effect was to have
6432 incorrect size information for certain domains.
6434 45. #include "config.h" was missing at the start of the crypt16.c module. This
6435 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6438 46. If there was a timeout during a "random" callout check, Exim treated it as
6439 a failure of the random address, and carried on sending RSET and the real
6440 address. If the delay was just some slowness somewhere, the response to the
6441 original RCPT would be taken as a response to RSET and so on, causing
6442 mayhem of various kinds.
6444 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6445 when I implemented it. It didn't allow for the fact that some option values
6446 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6447 the right test for positive values.
6449 48. Domain names in DNS records are case-independent. Exim always looks them up
6450 in lower case. Some resolvers return domain names in exactly the case they
6451 appear in the zone file, that is, they may contain uppercase letters. Not
6452 all resolvers do this - some return always lower case. Exim was treating a
6453 change of case by a resolver as a change of domain, similar to a widening
6454 of a domain abbreviation. This triggered its re-routing code and so it was
6455 trying to route what was effectively the same domain again. This normally
6456 caused routing to fail (because the router wouldn't handle the domain
6457 twice). Now Exim checks for this case specially, and just changes the
6458 casing of the domain that it ultimately uses when it transmits the message
6461 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6464 50. If a filter generated a file delivery with a non-absolute name (possible if
6465 no home directory exists for the router), the forbid_file option was not
6468 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6469 the existing equality matching.
6471 52. Exim was using ints instead of ino_t variables in some places where it was
6472 dealing with inode numbers.
6474 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6475 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6476 called TMPDIR, and if it finds it is different, it changes its value.
6478 54. The smtp_printf() function is now made available to local_scan() so
6479 additional output lines can be written before returning. There is also an
6480 smtp_fflush() function to enable the detection of a dropped connection.
6481 The variables smtp_input and smtp_batched_input are exported to
6484 55. Changed the default runtime configuration: the message "Unknown user"
6485 has been removed from the ACL, and instead placed on the localuser router,
6486 using the cannot_route_message feature. This means that any verification
6487 failures that generate their own messages won't get overridden. Similarly,
6488 the "Unrouteable address" message that was in the ACL for unverifiable
6489 relay addresses has also been removed.
6491 56. Added hosts_avoid_esmtp to the smtp transport.
6493 57. The exicyclog script was not checking for the esoteric option
6494 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6495 will work only if exicyclog is run under the appropriate euid.
6497 58. Following a discussion on the list, the rules by which Exim recognises line
6498 endings on incoming messages have been changed. The -dropcr and drop_cr
6499 options are now no-ops, retained only for backwards compatibility. The
6500 following line terminators are recognized: LF CRLF CR. However, special
6501 processing applies to CR:
6503 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6504 nor a local message in the state where . is a terminator.
6506 (ii) If a bare CR is encountered in a header line, an extra space is added
6507 after the line terminator so as not to end the header. The reasoning
6508 behind this is that bare CRs in header lines are most likely either
6509 to be mistakes, or people trying to play silly games.
6511 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6512 was being incorrectly given as 18 bytes larger than it should have been.
6513 This is a VOB (very old bug).
6515 60. This may never have affected anything current, but just in case it has:
6516 When the local host is found other than at the start of a list of hosts,
6517 the local host, those with the same MX, and any that follow, are discarded.
6518 When the list in question was part of a longer list of hosts, the following
6519 hosts (not currently being processed) were also being discarded. This no
6520 longer happens. I'm not sure if this situation could ever has previously
6523 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6526 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6527 header. This is something that is recommended in a new Internet Draft, and
6528 is something that is documented as being done by Sendmail. There are two
6529 possible values. For messages generated by the autoreply transport, Exim
6532 Auto-Submitted: auto-replied
6534 whereas for all other generated messages (e.g. bounces) it adds
6536 Auto-Submitted: auto-generated
6538 63. The "personal" condition in filters now includes a test for the
6539 Auto-Submitted: header. If it contains the string "auto-" the message it
6540 not considered personal.
6542 64. Added rcpt_include_affixes as a generic transport option.
6544 65. Added queue_only_override (default true).
6546 66. Added the syslog_duplication option.
6548 67. If what should have been the first header line of a message consisted of
6549 a space followed by a colon, Exim was mis-interpreting it as a header line.
6550 It isn't of course - it is syntactically invalid and should therefore be
6551 treated as the start of the message body. The misbehaviour could have
6552 caused a number of strange effects, including loss of data in subsequent
6553 header lines, and spool format errors.
6555 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6556 client host had authenticated. This control can now be exercised by an ACL
6557 for more flexibility.
6559 69. By default, callouts do not happen when testing with -bh. There is now a
6560 variant, -bhc, which does actually run the callout code, including
6561 consulting and updating the callout cache.
6563 70. Added support for saslauthd authentication, courtesy of Alexander
6566 71. If statvfs() failed on the spool or log directories while checking their
6567 size for availability, Exim confusingly gave the error "space shortage".
6568 Furthermore, in debugging mode it crashed with a floating point exception.
6569 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6570 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6571 serious problem, Exim now writes to the main and panic logs when this
6572 happens, with details of the failure. It then refuses to accept the
6573 incoming message, giving the message "spool directory problem" or "log
6574 directory problem" with a 421 code for SMTP messages.
6576 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6577 0, 1, and 2 exist, because some OS complain for execs without them (see
6578 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6579 descriptors. However, the code omitted to check that the open succeeded,
6580 causing mysterious errors if for some reason the permissions on /dev/null
6581 got screwed. Now Exim writes a message to the main and panic logs, and
6582 bombs out if it can't open /dev/null.
6584 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6585 interact so that it is all more flexible. It is supposed to remain
6586 backwards compatible. Also added extra_local_interfaces.
6588 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6589 to bomb out with an assertion failure - to the client this appears as a
6590 connection drop. This problem occurs in the part of the code that was taken
6591 from the Samba project. Fortunately, the assertion is in a very simple
6592 function, so I have fixed this by reproducing the function inline in the
6593 one place where it is called, and arranging for authentication to fail
6594 instead of killing the process with assert().
6596 75. The SPA client code was not working when the server requested OEM rather
6597 than Unicode encoding.
6599 76. Added code to make require_files with a specific uid setting more usable in
6600 the case where statting the file as root fails - usually a non-root-mounted
6601 NFS file system. When this happens and the failure is EACCES, Exim now
6602 forks a subprocess and does the per-uid checking as the relevant uid.
6604 77. Added process_log_path.
6606 78. If log_file_path was not explicitly set, a setting of check_log_space or
6607 check_log_inodes was ignored.
6609 79. If a space check for the spool or log partitions fails, the incident is now
6610 logged. Of course, in the latter case the data may get lost...
6612 80. Added the %p formatting code to string_format() so that it can be used to
6613 print addresses in debug_print(). Adjusted all the address printing in the
6614 debugging in store.c to use %p rather than %d.
6616 81. There was a concern that a line of code in smtp_in.c could overflow a
6617 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6618 initially expressed, the concern was not well-founded, because trailing
6619 spaces are removed early. However, if the trailing spaces were followed by
6620 a NULL, they did not get removed, so the overflow was possible. Two fixes
6623 (a) I re-wrote the offending code in a cleaner fashion.
6624 (b) If an incoming SMTP command contains a NULL character, it is rejected
6627 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6628 runs initgroups(), so that if the Exim user is in any additional groups,
6629 they will be used during message reception.
6635 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob